Systematics of the West Indian lizard genus Leiocephalus (Squamata: Iguania: Tropiduridae)

HE UNIVERSITY OF KANSAS MISCELLANEOUS
PUBLICATION

MUSEUM OF NATURAL HISTORY : No. 84

Systematics of the West Indian
Lizard Genus Leiocephalus
(Squamata: Iguania: Tropiduridae)

Gregory K. Pregill

LAWRENCE 30 April 1992

Museum of Comparative Zoology Library

Harvard University

THE UNIVERSITY OF KANSAS, MUSEUM OF NATURAL HISTORY
PUBLICATIONS

The University of Kansas Publications, Museum of Natural History, beginning with volume 1 in
1946, was discontinued with volume 20 in 1971. Shorter research papers formerly published in the
above series are now published as The University of Kansas Museum of Natural History Occasional
Papers. The University of Kansas Museum of Natural History Miscellaneous Publications began
with number | in 1946. Longer research papers are published in that series. Monographs of the
Museum of Natural History were initiated in 1970. Authors should contact the managing editor
regarding style and submission procedures before manuscript submission. All manuscripts are
subjected to critical review by intra- and extramural specialists; final acceptance is at the discretion
of the Director.

This publication is printed on acid-free paper. Occasional Papers and Miscellaneous Publications
are typeset using Microsoft® Word and Aldus PageMaker® on a Macintosh computer.

Institutional libraries interested in exchanging publications may obtain the Occasional Papers
and Miscellaneous Publications by addressing the Exchange Librarian, The University of Kansas
Library, Lawrence, Kansas 66045-2800, USA. Individuals may purchase separate numbers from the
Office of Publications, Museum of Natural History, The University of Kansas, Lawrence, Kansas
66045-2454, USA.

THE UNIVERSITY OF KANSAS

MUSEUM OF NATURAL HISTORY

MISCELLANEOUS PUBLICATION No. 84

30 April 1992

Systematics of the West Indian Lizard Genus
Leiocephalus (Squamata: Iguania: Tropiduridae)

GREGORY K. PREGILL

Department of Herpetology
San Diego Natural History Museum
PAOD Box 1590
San Diego, California 92112

MUSEUM OF NATURAL HISTORY
DycHE HALL
THE UNIVERSITY OF KANSAS
LAWRENCE, KANSAS

MISCELLANEOUS PUBLICATIONS

Editors for this issue: Richard F. Johnston and Linda Trueb
Managing Editor: Joseph T. Collins
Design and Typesetting: Kate A. Shaw and Joseph T. Collins

(or
LIBRARY
JUN 2 4 1992

HARVARD
WERE cr
M eous Publication No. 84
Pp. 1-69; 21 figures; 3 tables; 4 appendices
Published 30 April 1992
ISBN: 0-89338—041-5

© 1992 By Museum or Natura HIStory
DycHE HALL
THE UNIVERSITY OF KANSAS
LAWRENCE, KANSAS 66045-2454, USA

PRINTED BY
UNIVERSITY OF KANSAS PRINTING SERVICE
LAWRENCE, KANSAS

CONTENTS

ON FID COTS COS Ds 2 Ara CR PPE OEE 1
ANCTRINIO LTBI DY G3 0 SHIN ET DS Sate. core ae ae ESO a PA nee ae er ore re oR Semen POE l
NEN HES OANTE SuAINID NIE ERO ID S eee ee coset ict ccacsectece coe op aecccucmeneeee cceecee cuss cuesuacssesesceersssacsessssasasaasocses 2
LOGE Olle Wu RINNE AOA corre sock sade see orcneee eccececcemeceteteeteceeestsnets os Pubizes Sa ceneesnenneteassosssess 3
PHOCELHALUS MONOPHYLY AND OUTGROUP-REEATIONS HIP eore 2 ceecseseccncctccesouceeisesss 3
RIOTING) EIS <hosuaeececheor essecaconocscasetrocdo cone osce-cossce Renee socoaosoodacucaceosco4q0500 te Sdatod cr enib 060 9cF eso ganOEcseueeHoccoce >)

WM CTOGEP Malus INTOMOPINYLY cence srecct oscee-e-naset sees sa scsteessccnsceoaceonescareecesecetecs caceeses=arserta-taaeasecenaronresnsconse=- 6
GE NWA GREROA NAY SIS ‘ORWEEIO CEP ATU SS oooh scoscseess voce oe cess neie oe Seta es eee ont as as ece eae dceenaees 9
SS Ueet Neer tee ss rae eae eee sh cgn cece cae ean setsd svat Oe ees cee oaeo as taneses oben ace settrar cette ene eareee mec oeeraccaese se 9
UVa CU Le eee ee cre eet Sane ME Ned SATE aL, Demers ER Nee ne See c eet da desuaaudsed ners succes sossonscsenOsesesnese 14
OSECRAMIALES KEL SCOT eee reee cere se eee ae race ae a Saar a ane ee occ See Me aee eo tee te cana aoatasecescosssvoonsesussveesn eects 16
RCAC ES CALCS ee rcccscor i atc en acess he hac sanereestee tetas aehonaes tet POET Pe conte dn ERE on ee een cre Ca eS 16
ROVE S CALESIS Seeeat ee ccee cas ioca vec cacades aes nee ret neeer ee SO Pad EE on ree ee ee eee ees Pee rT COS 19
Silt [ERO] KG ee wake Uk oh i ON Er pe bd Oe ARR OA Re a mn rian erp sa een CRO er LEE RCUCCSO 20
(Ceasar inca Lil R221 1) oi Vag ema neon eee er ree een oP i Dr esiioree RN er rendre Bee Beni ronshatodosctnoG 21
(Ova STe. (CUBE Vel Cs cy Os ORR Re RARER EE Ee Re RC SSC cao ooen Sec ano ce Reeececec ae Cer eee Em pasaaarencaac ssococnc 22,
eee MOPOLOGIES FOR EEIO GEPHAUTUISS o.s.ccccestesttacteceeaiasenere nena dt oeee uceeee de Soe stdesmaaien sss aassonenees D2
EOS be Lj Ed OMG) Be 2 AWA GES oes, ee fo BTA, Vay PERERA geste ori peer ere PLEA AAD AE 29
WN Stalin Gli SS eee saree ee een Aut ph onal ein liek & ol) Ri oen. Sea neetee aa acaocathets cau sotae cher ose tew ans seen contonensenseeen® 29
GIVI LS OSS Seen ce ae eee nck ee ee saan ans c Sa ate teemeen concae a Sone race eae dae cn aceon TaSeeace Neen nee Nemtereenreeaees 29

New fossilsttnom the lbessen Amtiles”. sersc.cc. cc sessccseocecccsceesecesteeecoteccoassseseesteseeesessestacsteeceacceennc: Sil
IN@atinv AmmmeriGamy MeKtlaiyssccesessee-ceecteec ee ceeet cess scceeteesseet ence oes enone nono ssacresacsseacnetttaeatesrostenecsceess=easecas=: 33
ASCO) Gas GYAN Nena ee, cde ce ce eey Soe caw cslee tbe Pade SOaaee Rea a coca oes Nose e asiseec eau ttaa tas ave Seacooeoutencmeeeers 39)
SBE GUE Se NG COIN Siete cecccesreceteee tocar aoe cae eae ee cn ea Ue SESE seins retin s steps csaam erececautarncenresmsdadeslsseeeces 36
SUIS UN VAN RE it oe Rene be eek re RR stp d Aa SRR iS Ge ara nae ine PRP ARS Rree RDC er 53
POPU TM Eee ACTA) RE © TMT TD eee cee eece, cyte See stes ces toc cuccetee toseacmnccnmeeece enecesensiavasecess sasut ansaneananestnna sowessueenes 3)
/NTDIEVBINID) DC Es ati a ce AER Teac linet Rane Ree Fy PVA | eee ae er ee 59
PASE NID IEG Uli oe ae rete en ree ee Wa gt eg SAEED wee ceina ac Cnc erences Soe teins debe Seadgsasee gaccteaaet camtacessssseauccotcen 60
PSAP ES NIM UPC MTN Sn Sree ene eens eae sates see SESE NNO Ce Ne eRe ee ance ee are cag ua tvasauanene sac Atwacndnat cecussyeseweneenes 63

(STRAP BINT BDSG IN oo snore ieiada ss coonce J Soee conc Bi UsEeun er A SBC on apeae cee cree aeaa ecedarrroaccscoctcoBces soos B0cc6 67

a 7

>
s CKO ae
— SE Sete ec 4 7) _

a 7 : Ny a 7 7 7 a
ee Puebla, aera he
i 7 -
. a eee
- f

INTRODUCTION

In the arid landscapes of Cuba, Hispaniola,
and the Bahamas the swift, ground-dwelling liz-
ard Leiocephalus is often recognized by the sight
of its tail, which resembles a coiled watch spring.
The 21 extant species compose a relict assort-
ment now restricted to the West Indies north of
Puerto Rico (Schwartz and Thomas, 1975;
Schwartz and Henderson, 1988). Two species
that became extinct in the previous century and
six others known only from fossils reveal that the
range of this genus once included all of the major
islands and banks of the West Indies, at least as
far south as Martinique (Pregill et al., 1988).
They are small (50 mm SVL) to large (200 mm
SVL) lizards that sometimes are brightly col-
ored, but more often drab. Some have large scales,
others small. All are wary. Despite superficial
dissimilarities, lizards of the genus Leiocephalus
are a homogeneous lot.

Little is known about the natural history of
Leiocephalus or its relationship to other
neotropical squamates. A phylogeny of the spe-
cies never has been proposed, although consider-
able detailed information is available on their
geographic variation and distribution, and the
taxonomy of the group is reasonably stable ow-
ing primarily to documentation of the West In-
dian herpetofauna by Doris Cochran and Albert
Schwartz. In the late 1950’s Schwartz, inspired
by some of Cochran’s earlier (e.g., 1941) work,
initiated a series of studies on the various species

and populations of Leiocephalus that continued
for two decades. In the end, most of the living
species had been redescribed and, in some cases,
affinities among them were suggested.

A second major contribution to the systemat-
ics of Leiocephalus was restriction of the genus
to the Antillean species; since the previous cen-
tury, numerous South American taxa had been
included in the genus. However, Richard
Etheridge (1966a), who provided the first work-
able diagnosis of Leiocephalus, recognized that
evidence tying the Antillean forms with those
from the South American mainland was lacking
and recommended that the mainland species be
consigned to Ophryoessoides. During the past 15
years, several more West Indian species, both
living and extinct, have been named. Addition-
ally, a handful of fossils, mostly jaw fragments,
from the middle Tertiary of North America have
been referred to Leiocephalus (Estes, 1983).

The purpose of this study is to describe the
morphology of fossil and living Leiocephalus, to
test support for the monophyly of the genus, to
assess the taxonomic status of the putative North
American fossils, and based on an analysis of
morphological character transformations, to pro-
pose a phylogeny of the species. To facilitate
these ends, a species account containing the rel-
evant nomenclature and statements of distribu-
tion and characterization has been prepared for
each species. ;

ACKNOWLEDGMENTS

For the loan of specimens and/or access to
their collections I gratefully acknowledge the
following persons and their institutions: Richard
Zweitfel, Charles Myers, and George Foley, Ameri-
can Museum of Natural History (AMNH); E. N.
Arnold and Colin McCarthy, British Museum
(Natural History) [BM(NH)]; Jens Vindum, Rob-
ert Drews, Jacques Gauthier and Alan Leviton,
California Academy of Sciences (CAS and CAS-
SU); William Duellman, Linda Trueb, Darrel
Frost, and Michael Morrison, The University of
Kansas Museum of Natural History [KU and
ASFS (Albert Schwartz Field Series)]; Douglas
Rossman and Van Wallach, Louisiana State Uni-

versity Museum of Zoology (LSUMZ); Pere
Alberch and José Rosado, Museum of Compara-
tive Zoology, Harvard University (MCZ); Alain
Dubois and Ivan Ineich, Muséum National
d’Histoire Naturelle, Paris (MNHN); Richard
Etheridge, San Diego State University (REE, and
SDSU) Bruce MacFadden and Gary Morgan,
Florida State Museum, (UF); Arnold Kluge and
Gregory Schneider, University of Michigan Mu-
seum of Zoology (UMMZ); Michael Voorhies
and George Corner, University of Nebraska State
Museum (UNSM); Frederick Collier and Robert
Purdy, National Museum of Natural History
[USNM(VP)]; Ronald Heyer, George Zug, Roy

2 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

McDiarmid, Elyse Beldon and Ronald Crombie,
National Museum of Natural History (USNM);
William Presch, California State University, Ful-
lerton (WP). Numerous specimens in the herpe-
tological collections of the San Diego Natural
History Museum (SDSNH) also were examined.

Special thanks are extended to Donald Buden
for donating a fine series of Leiocephalus
loxogrammus that he collected on Rum Cay,
Bahamas, to Uno Svensson for information on
the Leiden specimen of Leiocephalus herminieri;
and to William Presch for technical assistance
with HENNIG86.

During various excursions to the West Indies,
my collecting and study of Leiocephalus were
enhanced immeasurably by the expertise and
good company of Ronald Crombie, Linda Gor-
don, Helen James, and Storrs Olson of the Na-
tional Museum of Natural History, David

Steadman of the New York State Museum, and
Richard Thomas of the University of Puerto Rico.
Without their participation and keen humor, along
with the support and cooperation of the
Smithsonian Institution, this study would have
been protracted far longer than it was.

Darrel Frost and Richard Etheridge shared
their ideas and insights during the course of
numerous discussions on iguanian systematics,
and made constructive comments on the manu-
script, which was also read critically by Ronald
Crombie, William Presch, Kevin de Queiroz, and
Ernest Williams. Any errors in logic, observa-
tion, or style are, however, my own. Figures 5, 6,
and 9 were drawn by Lynn A. Barretti.

This study was supported in part by grants
from the National Science Foundation (DEB-
8207347, BSR-8704630) and the National Geo-
graphic Society (No. 2482-82).

MATERIALS AND METHODS

Although the production of a cladogram 1s the
logical outcome of phylogenetic analysis, ex-
plicit descriptions of character attributes should
not be drafted impulsively in the desire to pro-
duce a tree. Character analysis is fundamental to
phylogenetic inference despite a growing body
of literature that is skewed heavily towards meth-
odology. Occasionally we are reminded that phy-
logenetic algorithms are most beneficial when
used to direct attention to trouble spots in the
data—viz., the characters themselves (e.g.,
Bryant, 1989).

In selecting characters, I sought attributes that
could be evaluated in two or more discrete con-
ditions, or “states.” Hence, attention was devoted
to uncovering variation in anatomical detail ow-
ing to size dependency or other causes that would
swamp a character’s discreteness and render it
ineffective for estimating relationships. Likewise,
I know of no reasoning that would suggest that
character analysis based on morphology should
not yield results (Kluge, 1989). Such potential
sources of information as biochemistry and kary-
ology also might prove to be revealing, but at
present, there are only isolated data of these kinds
available for Leiocephalus.

Any discrete morphological feature was con-
sidered a potential source of phylogenetic infor-
mation. There are numerous such features of the
skull, the postcranial skeleton, the integument,
musculature, and internal organs that have been
identified in squamates (e.g., Etheridge and de
Queiroz, 1988; Estes et al., 1988; Frost and
Etheridge, 1989). These and others were evalu-
ated for Leiocephalus and approximately 140
potential characters were examined across all
ingroup taxa. Of these, 39 proved sufficiently
discrete for use in phylogenetic resolution. The
justification for eliminating a potential transfor-
mation was simply confrontation with continu-
ous variation in that character (i.e., a lack of
discrete states), or more commonly because mean-
ingful variation was absent among species (1.e.,
all OTUs possess the same state).

After characters were identified and selected,
the direction of transformation (polarity) was
postulated according to the distribution of char-
acter states among hypothesized nearest rela-
tives, the first and second outgroups (Maddison
et al., 1984). When the polarity of a character
could not be ascertained confidently because of
ambiguous distribution of states among the

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 3

outgroups, that character was left unpolarized by
scoring the ancestral condition as unknown (a
*?”° in the data matrix). Likewise, multiple trans-
formations (those characters with more than two
states) could not always be ordered into additive
states. These were treated as “unordered” (= non-
additive) so as to allow the possibility for any
state to transform into another in only one step.

For ease of compilation, distributions of char-
acter states among the terminal taxa (= the spe-
cies) were analyzed with Swofford’s (1985, Ver.
2.4.) PAUP program, and Farris’s (1988, Ver. 1.5)
HENNIG86. PAUP has several options, of which
I used global swapping (SWAP = GLOBAL) and
multiple parsimony (MULPARS). Other utility
options include HOLD, which specifies the num-
ber of trees to be held in memory at each step of
tree construction. When HOLD = 1, only the

shortest, or one from the set of shortest, trees is
held for the next cycle. CSPOSS and BLRANGE
note character ambiguity on all but terminal stems,
and maximum and minimum branch lengths, re-
spectively. For the HENNIG86 analysis, I used
the following string of commands: MHENNIG*;
TPEOT BB; TPEOT; XSTEBS: HELM (Farris.
1988).

Characters were run without differential
weighting; that is, none was assumed to have
more phylogenetic meaning or content than any
other. Of course, the assumption that all transfor-
mations are equally likely predictors of phylog-
eny is, in itself, a form of weighting that might
not be true. But to do otherwise requires assump-
tions about development and evolution that I am
not prepared to defend.

CHOICE OF TERMINAL TAXA

For the ingroup analysis, I began with the
known species and subspecies of Leiocephalus
(Schwartz and Thomas, 1976; Schwartz and
Henderson, 1988) as terminal taxa. It was as-
sumed that these taxa were valid and diagnos-
able, although that assumption was tested in the
course of character analysis. One species of
Leiocephalus was unavailable—the Cuban en-
demic L. onaneyi known from three specimens
(Instituto de Zoologia [IZ], Academia de Ciencias
de Cuba; Garrido, 1973a). For all other living
taxa, both wet and skeletal specimens were ex-
amined, and, in certain cases, ample series of
males, females, and juveniles were available.
Some species are poorly represented in collec-
tions, perhaps known only from the type series or

a single skeleton that could be prepared from a
preserved series. The latter circumstance may
result in the erroneous assignment of a character
state to a species given that the range of variation
for a character is deduced from only one, or few,
individuals. A list of specimens examined is given
in Appendix IV.

As to fossils, those referred to Leiocephalus
do not provide the quantity of information com-
parable to that gleaned from whole specimens.
Therefore, the paleospecies are not included in
the primary data matrix. However, they are dealt
with in a separate section following the discus-
sion on ingroup topologies of living species and
there are analyzed together with the primary data
matrix.

LEIOCEPHALUS MONOPHYLY AND OUTGROUP RELATIONSHIP

In recognizing their shared attributes apart
from other Iguanidae, Etheridge (1966a) infor-
mally applied the name “tropidurines” to the
neotropical genera Liolaemus, Ctenoblepharys,
Phrynosaura, Stenocercus, Tapinurus, Plica,
Proctotretus, Uracentron, Uranoscodon, Tropi-
durus, and Leiocephalus, a group that also in-
cludes Phymaturus, Strobilurus, and Ophryo-

essoides. The “tropidurines” are one of eight
groups of iguanian lizards the monophyly of
which has been corroborated. However, “Iguani-
dae” in the traditional sense—all eight groups
comprising a family—remains unsubstantiated
as a natural taxon. There are no uniquely derived
attributes that all of these eight groups share, nor
is there more than suggestive evidence that would

4 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

relate one of these groups to any of the others
(Estes et al., 1988; Etheridge and de Queiroz,
1988; Frost and Etheridge, 1989; Williams, 1988).

The most recent phylogenetic analysis of
Iguania (“Iguanidae” + Agamidae + Chamae-
leonidae) by Frost and Etheridge (1989) failed to
uncover any evidence of a natural “Iguanidae.”
Hence, continued recognition of “Iguanidae” in
the formal sense perpetuates a concept of mono-
phyly when none exists. Frost and Etheridge
sought a logically consistent recourse to this
problem by proposing a revised taxonomy that
formalized the eight casual groupings into family
rank. The tropidurines of Etheridge (1966a), for
instance, become Tropiduridae. Within Tropidur-
idae, a formal subfamily, Tropidurinae, replaces
the casual “Stenocercus” group (Stenocercus +
Ophryoessoides + Proctotretus) plus the Tropi-
durus group (Plica, Uracentron, Uranoscodon,
Tropidurus, Strobilurus, and Tapinurus). In turn,
Liolaeminae formalizes the Liolaemus group
(Liolaemus, Ctenoblepharys, Phymaturus), which
thus leaves Leiocephalus as a monotypic sub-
family, the Leiocephalinae. The taxonomy of
Frost and Etheridge (1989) is followed here. For
a detailed analysis of the Tropidurus group, see
Frost (1987) and, in part, Rodrigues (1987), and
for the Liolaeminae, see Etheridge (MS).

The monophyly of tropidurids is supported by
possession of an enlarged sternal fontanelle, which
also occurs in phrynosomatids; however,
tropidurids differ in lacking femoral pores
(Etheridge, 1966a). The angular bone on the
mandible is reduced in tropidurids, as it is in
Anolis and other polychrids, and the gular fold is
incomplete medially as in Anolis and also the
phrynosomatid Sceloporus. Possibly, the
Tropiduridae can be distinguished by a unique
dentary-postdentary articulation (see below).

There is nothing compelling to suggest that
Leiocephalus is not nested within Tropiduridae,
but its relationship is arguable. When Etheridge
(1966a) restricted Leiocephalus to the West In-
dian species, he concluded that it was the least
easily placed tropidurine (= tropidurid) genus,
but that it seemed to have affinities with the large
(100+ species) South American complex Lio-

laemus. In more recent work, Frost (1987) and
Frost and Etheridge (1989) hypothesized that
Leiocephalus is the sister taxon of the “Stenocer-
cus’ + Tropidurus groups (Tropidurinae of Frost
and Etheridge, 1989).

I employed a suite of morphological charac-
ters extracted from the literature (Etheridge and
de Queiroz, 1988; Frost and Etheridge, 1989)
along with my own observations to evaluate al-
ternate outgroup relationships for Leiocephalus;
my results corroborated the phylogeny hypoth-
esized by Frost and Etheridge (1989) that is used
here (Fig. 1): Tropidurinae as the first outgroup
with Liolaeminae the second. Additional charac-
ter transformations that support this phylogeny
are described below.

Liolaemus group

Leiocephalus

Stenocercus group

rabies

——Uranoscodon

Tropidurus group

Fig. 1. Hypothesized relationships of Leiocephalus
with other Tropiduridae based on Frost and Etheridge
(1989) and this study.

1. Dentary-postdentary articulation (Tropi-
duridae).—The dentary-postdentary articulation
of tropidurids is unique among iguanians. The
articulation is formed primarily by an elongate
dentary that overlaps the surangular posterior to
the apex of the coronoid; the dentary and sur-
angular often fuse with one another in this region.
On the lingual side of the jaw, the antero-ventral
arm of the coronoid fits up underneath the medi-
ally produced dental shelf, flush against the
surangular internally. The dentary-postdentary
articulation of Anolis is similarly constructed,
but the dentary extends proportionately further
back onto the surangular. In iguanids, hoplo-
cercids, and corytophanids, the dentary scarcely
overlaps the surangular and the dentary-
postdentary articulation lies directly beneath the
coronoid, formed as a tongue-and-groove con-

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 3)

structed around an intramandibular septum that
descends from the roof of Meckel’s canal (Pregill,
1981). That of phrynosomatids is similar to the
iguanid joint, differing in the greater posterior
extension of the dentary onto the surangular.
Polarities of these various articulations are diffi-
cult to assess, and to order. Each may be derived
independently. For that reason I treat the tropidurid
condition only as a possible synapomorphy of the
family.

2. Palatal width (Leiocephalus + Tropiduri-
nae).—In pleurodont iguanians, the transverse
width of the palatal bones at the pterygopalatine
suture is nearly half or more the width of the
suborbital fenestra. In Leiocephalus and the
Tropidurinae, the pterygopalatine suture is nar-
row, being a third or less the width of the subor-
bital fenestra.

3. Quadrate (Tropidurinae).—In Leiocepha-
lus, numerous members of the “Stenocercus”
group, and most Liolaemus, the lateral conch of
the quadrate is broad, deep, and notched dorsally
to receive the peglike, quadrate process of the
squamosal. This is a basal squamate attribute
(Robinson, 1967; Estes et al., 1988). By contrast,
in most Tropidurus, Plica, Uranoscodon, Ura-
centron, Proctotretus, and some “Stenocercus”’
(humeralis, pectinatus, boettgeri, nigromaculatus,
praeornatus), the lateral conch of the quadrate is
shallow, nearly flat, and slightly or not at all
notched dorsally.

4. Infraorbital region (Tropidurus group).—
The palatine alone makes broad contact laterally
with the maxilla so as to separate the infraorbital
foramen from the lacrimal foramen in Tropidurus
(except west of the Andes), Plica, Uracentron,
Uranoscodon, Strobilurus, and Tapinurus. This
seems to be a synapomorphy of that group. In
other tropidurids (also Sceloporus), the lacrimal
bone is produced medially from the maxillary
arch to contact the palatine; thus, more of the
lacrimal and less of the palatine participates in
the bridge separating the infraorbital foramen
from the lacrimal foramen.

5. Fifth metatarsal (Tropidurus group).—
The primitive iguanian fifth metatarsal is planar
on the lateral side. In the Tropidurus group the
bone is convex laterally owing to the bone being

more robust overall and having a medial inclina-
tion of the distal process.

HoOMOPLASY

Frost and Etheridge (1989) discussed key in-
stances of homoplasy (with respect to Leio-
cephalus) in their cladogram of Tropiduridae.
For example, the nasal process of the premaxilla
is overlapped (to varying degrees) by the nasal
bones in Leiocephalus and the Liolaeminae; the
attribute is derived in these taxa, as is the posses-
sion of a coronoid labial blade. I note the follow-
ing additional convergences between Leiocepha-
Jus and other tropidurid taxa.

Narial foramen.—The anterior alveolar fora-
men located on the maxillary wall of the fossa
exonarina is enlarged in Leiocephalus, some
“Stenocercus, Plica, and Uranoscodon. The fo-
ramen is small or absent in other tropidurids, as
well as phrynosomatids and oplurids and, there-
fore, is considered the ancestral state.

Sternum size.—The sternum of Leiocephalus
(except herminieri) and the “Stenocercus” group
is small relative to that of other tropidurids and
most iguanians. Less than half of the sternum
extends posterior of the coracoids and the central
fontanelle comprises half or more the surface
area of the sternum. ;

Scapulocoracoid foramen.—The scapulocor-
acoid foramen, immediately dorsal to the glenoid
cavity, is conspicuously enlarged in Leiocephalus
and the “Stenocercus” group. In other tropidurids
and phrynosomatids, the foramen is small or
absent.

Number of distal tarsals.—There are two di-
stal tarsal elements in most squamates, numbers
1—3 having been lost. Leiocephalus and some of
the “Stenocercus” group (“Ophryoessoides” ca-
ducus, “O.” iridescens) have three distal tarsals,
as in Sphenodon; the additional element is visible
on the plantar surface proximal to the ends of
Metatarsals | and 2. Possibly the element results
from a secondary center of ossification in Distal
Tarsal 4. With some hesitation, I have assumed
that the additional element in Leiocephalus and
Ophryoessoides are homologues.

6 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

LEIOCEPHALUS MONOPHYLY

Monophyly of Leiocephalus is supported by
the derived attributes listed below. For the sake of
completeness, Table | provides a list of common
iguanian characters for which Leiocephalus ex-
hibits the ancestral state (Estes et al, 1988; Ethe-
ridge and de Queiroz, 1988; Frost and Etheridge,
1989).

Table 1. Plesiomorphic states exhibited by Leiocephalus

for common iguanian character transformations (e.g.,

de Queiroz, 1987; Etheridge and de Queiroz, 1988;
Frost and Etheridge, 1989; Williams, 1988).

1. Parietal foramen usually at frontal-parietal
suture
2. Nasal process of premaxilla narrow in some
3. Lacrimal present
4. Lacrimal foramen small
5. Postfrontal present
6. Supratemporal position usually lateral
7. Osseous labyrinth moderately evident
8. Splenial present, straplike
9. Angular present
10. Tooth crowns tricuspid, flared
11. Seven premaxillary teeth
12. Pterygoid teeth variously present
13. Palatine teeth absent
14. Second ceratobranchials short
15. Clavicles flat with moderate lateral flange
16. Clavicular fenestra absent
17. Median process of interclavicle long
18. Posterior coracoid fenestra absent
19. Twenty-four presacral vertebrae
20. Free ribs on lumbar vertebrae
21. Caudal autonomy present
22. Nuchal endolymphatic sacs absent
23. Scale organs smooth
24. Subdigital scale structure carinate
25. Distal subdigital scales without groove
26. Middorsal scale row present, continuous
27. Interparietal scale moderate
28. Superciliary scales elongate, strongly
overlapping
29. Single elongate subocular
30. Ulnar nerve superficial to limb muscle
31. Dorsal leg innervation from peroneal nerve
32. Nasal passage straight
33. Hemipenal muscles simple (Arnold, 1985)
34. Fourteen scleral ossicles

1. Nasals enlarged.—The nasals bones of Le-
iocephalus are large, being half again or more the
width of those in other tropidurids and
phrynosomatids.

2. Premaxilla.—In adult Leiocephalus, the
base of the nasal process of the premaxilla is
broad and, thus, the nasal process is triangular.
The shape is obtained ontogenetically from a
more uniformly tapered nasal process in juve-
niles (see Fig. 2 and Character 2, below).

3. Septomaxilla.—The septomaxilla of Leio-
cephalus is reduced in size and inflected ven-
trally at its posterior margin. In other tropidurids,
the septomaxilla is larger and the posterior mar-
gin is horizontal or directed dorsally. The re-
duced septomaxilla of Leiocephalus might be
correlated with the primitively short nasal vesti-
bule in these lizards (see Frost, 1987). However,
the ventral inflection posteriorly is peculiar to
them so far as I can ascertain.

4. Parietal roof.—tThe parietal roof of adult
tropidurids retains a neonatal, trapezoidal out-
line. In Leiocephalus, the sides of the parietal
table converge posteromedially into a “U” or a
“V” such that the adult parietal roof is more
triangular than trapezoidal (Fig. 3). Among
tropidurids, a convergent parietal table is pecu-
liar to Leiocephalus (Etheridge, 1966a). The con-
vergent parietal table, or a modification of it, also
obtains in most iguanids and in corytophanids,
with the latter group further expanding the “V”
into a posteriorly directed vertical blade (Lang,
1989):

5. Cervical rib morphology.—In tropidurids,
the cervical ribs of the fifth and sixth vertebrae
are variously expanded distally rather than proxi-
mally, as is the case in most other Iguania. In
numerous Liolaemus, Phymaturus (patagonicus)
and a few members of the “Stenocercus” group
(crassicaudatus, caducus), the ribs are nearly
twice as wide distally as they are proximally. In
the Tropidurus group, the cervical ribs are shaped
like dumbbells, being expanded proximally and
distally. Leiocephalus is peculiar in that the cer-
vical ribs are strongly recurved, rather than
straight, and expanded in the middle. In the ab-
sence of a discernable morphocline, I would treat
the transformations in shape (from the primitive,
tapered condition) as separate characters—greatly

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS al

Nasal

Fig. 2.
Overlap may be minimal (left, Leiocephalus inaguae UMMZ 149133), leaving the nasal process exposed beyond
the posterior border of the external nares or extensive (right, L. barahonensis SDSNH 64578) so that the process
is obscured almost completely (Character 1).

expanded distally or not, and dumbbell shaped or
not. The scooped, recurved condition of Leio-
cephalus is an autapomorphy of the genus.

6. Caudal vertebrae.—There are two apomor-
phic features of Leiocephalus caudal vertebrae.
One is the more proximal occurrence of the first
caudal vertebra bearing an autonomic fracture
plane (characteristically the fifth or sixth verte-
bra from the sacrum). In some “Stenocercus” and
some Liolaemus (also some Sceloporus), the first
fracture plane usually occurs at the seventh ver-
tebra, whereas in all other iguanians, it is located
at the eighth or beyond. Perhaps the more ante-
rior placement of caudal autonomy in Leio-
cephalus is correlated with their tail-curling be-
havior (see below). However, in another
tail-waver, Callisaurus, the most anterior verte-
bra with a fracture plane is usually posterior of
the eighth.

The second autapomorphy is the presence of a
short, dorsally projecting spike located directly
above the autonomic fracture plane (Etheridge,
1966a).

7. Interclavicle.—In most Scleroglossa, the
lateral processes of the interclavicle diverge from
the anterior terminus of the median process and

Premaxilla

Overlap of the nasal process of the premaxilla (premaxillary spine) by the nasal bones in dorsal aspect.

contact (or are overlain by) the proximal rami of
the clavicles. In Leiocephalus, the interclavicle
bears a short, median process anterior to the
lateral processes—i.e., the interclavicle has an
‘anterior process” (Fig. 4). Thus, the lateral pro-
cesses are displaced posteriorly and are free of
the clavicles. Also, the posterior process of the
interclavicle is broadly flared-

8. Xiphisternal rods.—In Leiocephalus, the
xiphisternal rods continue posteriorly beyond the
last (second) pair of xiphisternal ribs, whence
they curve back anteriorly beneath the last pair of
xiphisternal ribs (Etheridge, 1966a; Fig. 4). A
parallel condition occurs in Tapinurus, but the
recurved rods continue ventrally in association
with the M. pectoralis (Frost, 1987).

9. Nasal and rostral scale contact.—Only
Leiocephalus among tropidurids has the nasal
scales in broad contact with the rostral scale.
Apparently, Leiocephalus has lost the postrostral
scales that separate the nasals from the rostral in
other tropidurids and most other iguanians.

10. Enlarged cephalic scales.—The head
scales (frontonasals, frontals, supraoculars, pari-
etals) of Leiocephalus are exceptionally large
relative to those of other tropidurids (Fig. 11).

8 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Rigas:
the parietal table in adults becomes either V-shaped (left) or U-shaped (right) (Character 8).

Only among some species of “Ophryoessoides”
are supraocular scales so enlarged, but not to the
extent characteristic of Leiocephalus.

11. Lenticular scale organs.—On the trailing
edge of the paravertebral, ventral, and caudal
scales of Leiocephalus, there are from two to five
lenticular organs on either side of the median
keel; the tip of the keel also bears a terminal scale
organ (Etheridge, 1966a:fig. 9a). One, or occa-
sionally two, organs are found on the paravertebral
scales of “Stenocercus” crassicaudatus, some
Tropidurus (e.g., T. stolzmani) and “Stenocercus”
apurimicus. In “Ophryoessoides” irredescens,
“O.” caducus, and species of Liolaeminae, there
may be a terminal lenticular scale organ on the

The ontogenetic convergence of the parietal table in dorsal view. From a flat, juvenile condition (top)

keel. The presence of multiple scale organs on
most, or all, body scales and on the tail is appar-
ently a derived feature of Leiocephalus.

12. Tail curling.—As far as known, most spe-
cies of Leiocephalus curl the tail in display, a
behavioral trait unique among tropidurids. Sweep-
ing the tail upward in a high spiral is performed
by both sexes of L. carinatus coryi during court-
ship and territorial defense (Evans, 1953). Also,
I have witnessed tail-curling in L. barahonensis,
carinatus, loxogrammus, lunatus, macropus,
personatus, raviceps, schreibersi, and
semilineatus. Possibly, the behavior is not prac-
ticed by pratensis (fide observations by Richard
Thomas published in Schwartz, 1968).

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 9

CHARACTER ANALYSIS OF LEIOCEPHALUS

SKULL

Numerous details of the skull were evaluated
as potential transformations, but few could be
characterized without insurmountable problems.
Most structural details do not vary between spe-
cies, but some vary intraspecifically and others
are size-dependent. Much of this variation is
described under the transformations enumerated
(1-11) below. For reference to its general mor-
phology, the skull of Leiocephalus carinatus il-
lustrated in Figures 5 and 6, is characteristic.

Fig.4. The pectoral girdle of Leiocephalus, ventral
view.

In Leiocephalus, snout profiles are neither
strongly vaulted nor obviously depressed, as they
are in some other tropidurids. Drawings of lateral
views of the skull were used to measure the angle
formed by the nasal process of the premaxilla and

the maxillary tooth row. The angle ranges from
42—55° as a continuous, ontogenetic and indi-
vidual variable, thereby precluding any mean-
ingful partitioning into discrete states. To this,
the following can be added: the number and
location of maxillary foramina vary individually;
interorbital width becomes proportionately
broader with increase in skull size; among spe-
cies, the vomers, palatines, pterygoids, and
ectopterygoids are similar to one another in shape
and in articulations; pterygoid teeth are variably
present within and among species; variation in
the width of the interpterygoid vacuity is largely
an artifact of skeletal preparation and skull size;
the posterior processes of the basisphenoid vary
in their encroachment onto the sphenooccipital
tubercles, but variation is inconsistent within a
species; other details of the basicranium are re-
markably consistent from one species to the next.

I have attempted to interpret characters con-
servatively in order to minimize ambiguity in
character-state assignment.

1. Nasal overlap of premaxilla (Fig. 2).—In
Leiocephalus and the Liolaeminae, the nasal pro-
cess of the premaxilla (hereafter referred to as the
“premaxillary spine’) is enveloped dorsally and
ventrally by the nasal bones. In all other
tropidurids, the premaxillary spine overlies the
anteromedial confluence of the nasals, as in many
squamates. In Liolaemus, the nasal bones may
overlap the distal end of the premaxillary spine,
characteristically that portion posterior to the
external nares. Within Leiocephalus, increasing
nasal overlap accompanies ontogeny in most spe-
cies, such that in adults, the nasals cover as much
as two-thirds of the spine. Increasing overlap of
the premaxillary spine (State 1) is characteristic
of all Leiocephalus except for eight species in
which the nasals fail to converge completely over
the distal half (State 0): greenwayi, herminieri,
inaguae, loxogrammus, punctatus, psammo-
dromus, some carinatus, and some /unatus.

2. Premaxillary shape (Fig. 2).—The shape
of the nasal process of the premaxilla varies with
size, being nearly parallel-sided in small indi-

10 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Fig. 5. The skull of Leiocephalus carinatus varius (USNM 217299) in dorsal and ventral views.

Fig.6. The skull of Leiocephalus carinatus varius (USNM 217299) in lateral view.

viduals, but triangular in larger forms. However,
one discrete feature is discernable. In adult L.
inaguae, macropus, punctatus, psammodromus,
and schreibersi, the base of the premaxillary
spine is constricted just dorsal to the dentigerous

process (State 1). In all other species, including
outgroup taxa, the sides of the premaxillary spine
are parallel or evenly tapered (State 0)

3. Premaxillary spine, lateral spike (Fig.
2).—In inaguae and psammodromus, there is a

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 1]

short, laterally projecting spike on either side of
the premaxillary spine immediately dorsal to the
basal constriction (State 1). A lateral spike is
absent in other Leiocephalus and the outgroups
(State 0).

4. Nasal-maxillary suture.—The nasal bones
of Leiocephalus form the dorsal and lateral walls
of the bony external nares. The anterolateral
processes of the nasals curve and taper down the
anterior side of the nasal process of the maxilla,
but the extent of that overlap varies among indi-
viduals. Dorsally, on the snout, nasal articulation
with the maxilla in most Leiocephalus is charac-
teristic of other tropidurids—1.e., the nasal-max-
illary suture arcs toward the midline of the snout
as a consequence of the dorsomedial extension of
the nasal process of the maxilla (State 0). In L.
lunatus, rhutidira, semilineatus, and vinculum
endomychus, the nasal-maxillary suture is nearly
straight-sided, exhibiting minimal convergence
toward the midline of the snout (State 1).

5. Nasal processes of frontal exposure (Fig.
7).—In tropidurids, the prefrontal abuts the nasal
bone for half or more the length of the lateral
margin of the nasal bone itself. The nasals and
prefrontals contact one another and thereby ob-
scure the nasal processes of the frontal bone
(State 0). Occasionally, the nasals and prefron-
tals do not meet posteriorly and the nasal pro-
cesses of the frontal are exposed between them
(State 1). The extent of nasal-prefrontal contact
may vary within species, particularly among ju-
veniles and subadults in which reduced contact is
seen more often. Nonetheless, the pattern is suf-
ficiently consistent to characterize (Fig. 7). Re-
duced nasal-prefrontal contact in adults occurs in
L. greenwayi, herminieri, loxogrammus, macro-
pus, punctatus, rhutidira, vinculum endomychus,
some carinatus, and some personatus. Because
both states occur among the “Stenocercus” group
(crassicaudatus, guentheri, rhodomelas, irides-
cens), Plica, Urocenteron, and the Liolaeminae,
the polarity of this character is equivocal (ances-
fon—s)):

6. Septomaxilla.—As characters, septomaxil-
lae are treacherous to evaluate because of their
comparatively small size, entrenchment within
the nasal chamber, and the fact that they are

delicate bones easily damaged in skeletal prepa-
ration. Few are preserved in series sufficient to
provide confident description and comparison.
The following transformation, therefore, is of-
fered cautiously.

The septomaxilla of Leiocephalus is a small,
saddle-shaped bone that follows the plane of the
premaxillary spine proximally, and then curves
ventrad toward the vomer, rather than continu-
ing anterodorsally in the direction of the pre-
maxillary spine. The bone is similar to that of
other tropidurids in bearing a square to rectan-
gular lateral wing with a thin, posteriorly di-
rected process protruding freely into the nasal
cavity (State 0). In L. herminieri, loxogrammus,
melanochlorus, and psammodromus, the lateral
wing and posterior process are reduced or ab-
sent (State 1).

7. Frontal, posterior width.—In some Leio-
cephalus, the posterior half of the frontal is un-
usually wide and flat, and the interorbital region
is less narrow and furrowed (less concave) than
in other species (State 1). A narrow frontal is
primitive for tropidurids (State 0). Admittedly,
“width” is subjective, and length/width ratios of
adult frontals do not reveal an obvious difference
from what I would consider narrow or wide fron-
tals in Leiocephalus. Nonetheless, visual com-
parison was sufficiently compelling to score as
derived the wide, flat frontal. of inaguae, macro-
pus, melanochlorus, pratensis, psammodromus,
and punctatus.

8. Parietal table, convergence (Fig. 3).— All
Leiocephalus undergo an ontogenetic vaulting of
the parietal table (Etheridge, 1966a). In juve-
niles, the parietal table is comparatively broad
and flat, not unlike that of other tropidurids as
adults. With growth, the lateral faces of the pari-
etal steepen and converge posteromedially to-
ward one another as araised ridge that is U-shaped
in some species (State 0) and V-shaped in others
(State 1). Neither seems to correlate directly with
absolute size; however, the V-ridge seems to be
the more derived transformation ontogenetically.
Additionally, in taxa having the V-ridge, the
supratemporal processes diverge from the pari-
etal table at a more acute angle. The V-shaped
ridges and associated acutely diverging supra-

igi.

Frontal

Prefrontal

2 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Nasal

Dorsal view of nasal-frontal region of the skull of Leiocephalus. The nasal processes of the frontal bone

may be obscured by confluence of the nasals and prefrontals (left) or remain exposed (Character 5).

temporal processes characterize L. greenwayi,
inaguae, loxogrammus, psammodromus, and
punctatus.

9. Supratemporal, position.—In most Leioce-
phalus, the supratemporal bone lies on the lateral
side of the supratemporal process of the parietal,
as it does in tropidurids and squamates generally
(State 0; Etheridge and de Queiroz, 1988; Estes et
al., 1988). In L. barahonensis, semilineatus, v.
vinculum, psammodromus, and some loxogram-
mus (1. parnelli), the supratemporal occupies a
ventromedial position (State 1).

10. Squamosal (Fig. 8).—In most Leiocepha-
lus and the “Stenocercus” and Tropidurus groups,
the posterior end of the squamosal is expanded
dorsally as a broad ridge that continues onto the
anterior ramus of the squamosal. The dorsal
(supratemporal) process, distinct in many
iguanians, is nearly confluent with the expanded
anterior ramus and is no longer evident (State 0).

In L. psammodromus, melanochlorus, and some
loxogrammus (1. loxogrammus) and macropus,
the proximal end is not expanded; thus, the ante-
rior ramus is narrow and the supratemporal pro-
cess 1s distinct (State 1).

11. Skull rugosities—The dermal roofing
bones (parietal, frontal, nasals) of L. barahonensis,
some carinatus, cubensis, greenwayi, herminiert,
lunatus, personatus, raviceps, rhutidira, semi-
lineatus, and stictigaster are sculptured with an
irregular pattern of prominent rugosities (State
1). In larger individuals sculpturing may include
the impressions of the overlying head scales.
Smooth roofing bones are characteristic of all
other species and most outgroup taxa (State 0)
with the exception of some individual variants of
“Stenocercus” and “Ophryoessoides” (e.g., O.
aculeatus, O. iridescens, S. apurimacus), and
apparently some Liolaemus (Etheridge and de
Queiroz, 1988).

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS

Squamosal

Fig. 8. The squamosal in lateral view. The bottom figure depicts the bone with
a peglike supratemporal process that is characteristic of most Leiocephalus and
numerous other iguanians. The squamosal of other Leiocephalus and Tropidurinae
(top) lacks a well-defined supratemporal process (Character 10).

14 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

MANDIBLE

Numerous characters of the mandible have
been identified in iguanian taxa and applied with
varying degrees of consistency and success in
phylogenetic analyses (de Queiroz, 1987; Frost,
1987; Estes et al., 1988; Etheridge and de Queiroz,
1988; Frost and Etheridge, 1989). For Leio-
cephalus (Fig. 9), | evaluated length of coronoid
overlap on the anterolabial side of the dentary;
length of coronoid on the anterolingual side of
the dentary; shape and height of coronoid; posi-
tion and/or confluence of the anterior inferior
alveolar foramen and anterior mylohyoid fora-
men; anterior and posterior extent of splenial;
position of posterior mylohyoid foramen; rela-
tionship of anterior supra-angular foramen with
coronoid; number of mental foramina; shape and
position of angular; shape and orientation of the
retroarticular process; number of teeth, and shape
of tooth crowns. None of the aforementioned
occurs as characters divisible into discrete states.
Two, however, could be scored—emargination
of the posterior border of the dentary, and the
transition from simple to tricuspid tooth crowns.

Fig. 9.

12. Dentary emargination (Fig. 10).—The
posterior end of the dentary bears a dorsal
surangular process and a ventral angular process,
between which is a notch or emargination. Pos-
session of an emarginate dentary may be a
synapomorphy of Leiocephalus, because among
other tropidurids it occurs only in some eastern
Tropidurus (Darrel Frost, pers. comm.). Primi-
tively, the back of the dentary tapers postero-
dorsally so that only a surangular process exists.
An emarginate dentary occurs in iguanids and
corytophanids, but homology with Leiocephalus
is doubtful because of the dissimilar construction
of the dentary-postdentary articulation (see
above).

Among Leiocephalus, the angular process is
developed to varying degrees, from quite short to
long, being equivalent to the surangular process
in size. For this analysis, variation was collapsed
into two states: (1) angular process not well
developed (less than half the size of the surangular
process) and therefore only modest emargination
of the dentary is evident (State 0); or (2) angular
process pronounced (at least half or more the size

Left mandible of Leiocephalus carinatus varius (USNM 217299) in medial (top) and lateral views.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 15

Fig. 10. Emargination of the posterior end of the dentary in some Leiocephalus results froma prominent angular
process produced below a dorsal surangular process as in the bottom figure (L. personatus USNM 225041). In other
species, the angular process is rudimentary and only the surangular process is evident (top, L. psammodromus
UMMZ 149109) (Character 12).

of the surangular process), and having obvious personatus, raviceps, rhutidira, schreibersi,
emargination of the dentary (State 1) asis found — semilineatus, stictigaster, and vinculum.
in L. barahonensis, cubensis, loxogrammus, 13. Transition to tricuspid tooth crowns.—

16 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

The transition from simple, unicuspid tooth
crowns to tricuspid crowns takes place anteriorly
at the third or fourth tooth in Leiocephalus
rhutidira and L. vinculum endomychus, and as far
posteriad as the tenth or eleventh tooth in all
other species except L. greenwayl, inaguae,
loxogrammus parnelli, and lunatus, in which the
transition occurs between the seventh to ninth
teeth. Because the transition becomes more
posteriad ontogenetically (i.e., as the tooth row
lengthens), I evaluated this character in series of
dentaries with comparable tooth row lengths (ca.
8-10 mm). A survey of outgroup species re-
vealed that the transition to tricuspid crowns may
occur on Teeth | or 2 (“Stenocercus” roseiventris),
5-7 (Stenocercus apurimacus, Plica plica), 7-9
(Urocentron, Stenocercus boettgeri) and 10 or 11
(“Ophryoessoides” iridescens, Stenocercus
humeralis, Tropidurus peruvianus, Uranoscodon,
and Sceloporus occidentalis). Insome Liolaemus,
all premaxillary and maxillary teeth are tricuspid
(Richard Etheridge, pers. comm.), whereas in
others, the transition from unicuspid to tricuspid
crowns takes place at the fifth tooth or beyond.
For this analysis, I consider only the most ante-
rior transition (Teeth 3 or 4) in Leiocephalus as
State | and all other conditions as State 0, but the
character is left unpolarized.

POSTCRANIAL SKELETON

14. Vertebral neural processes.—In Leio-
cephalus, the neural spines of the midbody verte-
brae are vertical and expanded distally along the
longitudinal axis. This is ancestral for tropidurids
(State 0). In some L. carinatus and loxogrammus
(1. parnelli), and in raviceps and schreibersi, the
spines are low, obtuse, and not distally expanded
(State 1).

15. Vertebral hypapophyses.—The hypapo-
physes of the trunk and lumbar vertebrae (exog-
enous outgrowths of the centra) of tropidurids are
flat, uniformly broad, and approximately half the
width of the condyles of the centra (State 0). In L.
loxogrammus, personatus, raviceps, and some
lunatus, the hypapophyses are narrow, longitudi-
nal ridges (State 1).

Girdles.—The pectoral girdle of Leiocephalus
(Fig. 4) was scrutinized, but other than generic

autapomorphies (see above), no characters were
discovered. Because clavicle geometry has been
applied at other levels of analysis in iguanians, it
is worth noting the ontogenetic change in this
bone in Leiocephalus; the angle formed by the
proximal and distal rami increases concomitant
with a broadening of the bone posterolaterally.
The clavicle of L. schreibersi is unique in being
more slender distally and less acute overall than
that of other species.

As with the shoulder girdle, the pelvis of
Leiocephalus is a generalized structure and no
usable characters were found.

HEAD SCALES

Most of the primary descriptions and diag-
noses of extant species of Leiocephalus (see Spe-
cies Accounts) rely exclusively on squamation of
the head and body. Commonly, this includes
meristic values (means and range) for prefron-
tals, loreals, temporals, supraoculars, supraocular
semicircles, dorsal crest scales from occiput to
vent, and tricarinate scales on the fourth toe.
Although modal differences in scale counts are
useful in differentiating species, the values are
broadly overlapping and difficult or impossible
to characterize and assign polarity.

The enlarged cephalic scales of Leiocephalus
are the most distinctive feature of their squama-
tion. Among tropidurids, the enlargement of head
scales (parietals, supraoculars, prefrontals) is
approached only by a few species of “Ophryo-
essoides.” The head scale patterns shared among
Leiocephalus and described below (Characters
16 and 18) are unique.

16. Snout scales (Fig. 11)—Among Leioce-
phalus, scales of the nasal and prefrontal region
(the frontonasals and prefrontals of Smith, 1946)
are arranged in one of three patterns. For conve-
nience I designate these patterns as Types I, I,
and II. In the Type I pattern, there are three or,
rarely, four transverse rows of subequal scales
between the internasals and the anterior pair of
frontals (the latter = median head scales of
Schwartz, 1967a). None of these scales is en-
larged, as they are in Type II and ITI patterns, and
there are at least four scales in contact with the
anterior pair of frontal scales (State 1). The Type

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 7

O58

Fig. 11.

Patterns of frontonasal and parietal scales. The three patterns of snout scales (horizontal hatching) in

Leiocephalus result from a decrease in the number of frontonasals and the concomitant enlargement of those that
remain. From left to right: Type I (L. psammodromus USNM 30385); Type II (L. stictigaster USNM 140466); Type
Il (L. barahonensis SDSNH 64582) (Character 16). Three patterns of parietal scale patterns (shading). From left
to right: Type I, lateral parietals are smaller than medial pair; Type II, lateral parietals are equal to medial pair; Type
III, lateral parietals are larger than medial pair (Character 18).

I pattern characterizes L. eremitus, herminieri,
melanochlorus, and psammodromus.

The Type II pattern is a configuration of three
rows of scales between the internasals and ante-
rior frontals. The middle row may include a pair
of enlarged scales (Character 17), whereas the
posterior row is composed of three smaller scales
in contact with the anterior pair of frontals (State
2). This pattern is characteristic of L. carinatus,
cubensis, greenwayi, inaguae, macropus, punc-
tatus, Schreibersi, and stictigaster.

The Type III pattern is distinguished by two
rows of scales between the internasals and the
anterior pair of frontals; the posterior row is
composed of three, or a single pair of, enlarged
scales in direct contact with the anterior pair of
frontals (State 3). The Type II pattern occurs in
L. barahonensis, loxogrammus, lunatus, person-
atus, pratensis, rhutidira, raviceps, semilineatus,
and vinculum.

None of these patterns is repeated among
outgroup species (State 0), and although each
could represent an independent derivation, the
patterns seem to be related serially with the con-

dition of more numerous, smaller scales leading
to one with fewer, larger scales (I — IH — III).
Nonetheless, this transformation is left unordered.

17. Frontonasal scales, enlarged pair.—
Three species have a greatly enlarged median
pair of frontonasal scales (State | )—Leiocephalus
carinatus, greenwayi, and punctatus. In all other
species, these scales are not enlarged (State 0).

18. Parietal scales (Fig. 11).—There are four
large parietal scales in Leiocephalus—a lateral
pair and a median pair. Immediately posterior to
the parietals there may be up to three irregular
rows of small, postparietal scales occupying the
nuchal fold (Character 33). As with the snout
scales, there are three discernable patterns of
parietal scales, which are designated as Types I,
II, and III.

Overall, the parietal scales of Type I are smaller
than those of either Type I or III; the most lateral
parietal scales are smaller than the median pair, and
there are two to four irregular rows of postparietal
scales present (State 1). Type-I species include
Leiocephalus herminieri, inaguae, macropus,
melanochlorus, psammodromus, and schreibersi.

18 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

The Type-III pattern is extreme with respect to
Type I. The most lateral parietal scales are larger
than the median pair, rather than smaller, and all
four scales are, overall, larger than those of Types
I and II. Postparietal scales are few, more often
absent (State 3). The Type III pattern character-
izes Leiocephalus barahonensis, carinatus,
eremitus, greenwayi, pratensis, punctatus, and
vinculum altavelensis.

The Type-II pattern is intermediate to Types I
and III; the most lateral parietal scales are subequal
to, or slightly larger than, the median pair, and
there is a single row (occasionally two), of
postparietal scales (State 2). Type-II species are
Leiocephalus cubensis, loxogrammus, lunatus,
personatus, raviceps, rhutidira, semilineatus,
stictigaster, and vinculum.

An appeal to outgroups is inconclusive with
respect to the direction of transformation, and
this character is left unordered. Lack of any
conspicuously enlarged parietal scales is treated
as the plesiomorphic condition (State 0).

19. Lateral postparietal scale, enlarge-
ment.—An apomorphy of Leiocephalus lunatus,
loxogrammus, raviceps, and semilineatus is the
presence of an enlarged postparietal scale on
either side of the head, adjacent to each lateral
parietal (State 1). These scales are not enlarged in
other species or outgroups (State Q).

20. Internasal scales (Fig 11).—Most
Leiocephalus are unique among tropidurids in
having fewer than four internasals, and in having
nasals that contact the rostral scale; presumably
the postrostral scales have been lost in Leio-
cephalus. Only L. herminieri and some psammo-
dromus retain four internasals. In L. macropus
and most (70% of specimens) pratensis and
melanochlorus, there are two internasal scales,
whereas all other species have three. All three
internasals usually contact the rostral in L.
carinatus, greenwayi, loxogrammus, eremitus,
and some /unatus (1. arenicolor), but in other
species with three internasals, usually only two
are in broad contact with the rostral, with the
median internasal being smaller and posteriorly
displaced. Because of considerable intraspecific
variation in broad internasal contact with the
rostral (i.e., 2 vs. 3 scales), I have simplified this

transformation as follows: four internasals (State
()), three internasals (State 1), or two internasals
(State 2).

21. Lorilabial scales (Fig. 12).—In Leioce-
phalus, the lorilabial scale row (that row imme-
diately dorsal to the supralabial scales) consists
of four to eight scales. The two most posterior
lorilabials are bordered above by an elongate
subocular. In Tropidurinae and most Liolaeminae,
there are three or four lorilabial scales anterior to
the first lorilabial in contact with the elongate
subocular. Most species of Leiocephalus also
have four (State 0) except L. eremitus, melano-
chlorus, psammodromus, and punctatus. These
species have five or six lorilabials anterior to the
first lorilabial scale contacting the elongate
subocular (State 1).

eae
A,

Flees lis
barahonensis (SDSNH 64582), showing the enlarged
temporal scale (vertical hatching) anterodorsal to the ear
(Character 23).

Lateral head scales of Leiocephalus

22. Cephalic scale ridges.—Many Leiocepha-
lus possess multiple series of longitudinal ridges
on the parietal scales, supraoculars, frontals and,
in some species, most of the scales of the snout.
“Ophryoessoides” caducus and Proctotretus have
keeled head scales, but these are doubtfully ho-
mologous to the low, parallel ridges of Leioce-
phalus. Smooth head scales otherwise character-
ize the “Stenocercus” and Tropidurus groups and,
on that basis, scale ridges in Leiocephalus are
treated as apomorphic. The cephalic scales of L.
barahonensis, carinatus, eremitus, herminieri,
inaguae, greenwayi, psammodromus, punctatus,
rhutidira, semilineatus, and vinculum are smooth

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 19

or bear indistinct ridges posteriorly (State 0).
Species with ridges restricted to the parietals,
supraoculars and frontals (State 1) are L.
loxogrammus, lunatus, personatus, pratensis, and
schreibersi, and those with well-defined ridges
extending onto the frontonasals (the most de-
rived condition, State 2) of this three-step trans-
formation, are L. cubensis, macropus, melano-
chlorus, raviceps, and stictigaster.

23. Temporal scale enlarged (Fig. 12).—The
temporal scales of most Leiocephalus are
subequal, as they generally are in Tropiduridae
(State 0). Dunn (1920) described the single, en-
larged temporal scale anterodorsal to the ear in L.
semilineatus, which Schwartz (1967a) also dis-
covered in L. barahonensis (State 1; Fig. 12). The
enlarged temporal scale is not always present in
L. semilineatus (absent in about 25% of my
sample) and it occurs with about the same fre-
quency in L. rhutidira. To complicate matters, a
“moderately” enlarged temporal scale occurs in
some individuals of L. /unatus, punctatus, vincu-
lum (except v. vinculum), and less occasionally,
in L. personatus. A number of Liolaemus also
possess an enlarged temporal scale (R. Etheridge,
pers. comm.), but homology with Leiocephalus
is unclear. Because the enlarged scale of the first
three species always is located at the anterodorsal
corner of the auricle, but may be more dorsal or
anterior in the others, I have restricted the de-
rived state to L. rhutidira, semilineatus, and bara-
honensis.

24. Temporal scales elongate.—Another
peculiarity of temporal scalation is the presence
of two or three elongate scales immediately be-
hind the eye in Leiocephalus loxogrammus and
L. raviceps (State 1). No such elongate temporals
occur in other species or outgroups (State 0).

Bopy SCALES

25. Neck scales, lateral —Among Leiocepha-
lus, the lateral scales of the neck are either keeled
and undifferentiated with respect to the surround-
ing body scales (State 0), or they are smaller, and
more granular with keeling reduced or absent
(State 1). Neither state correlates directly with
the elaboration of skin folds in Leiocephalus (see

below). Undifferentiated neck scales are found in
L. carinatus, barahonensis, cubensis, lunatus,
personatus, loxogrammus, pratensis, raviceps,
rhutidira, semilineatus, stictigaster, and vincu-
lum, whereas small lateral neck scales occur in L.
eremitus, greenwayi, herminieri, inaguae,
macropus, melanochlorus, psammodromus,
punctatus, and schreibersi.

Outgroup criteria are equivocal with regard to
polarity, because both differentiated and
undifferentiated lateral neck scales occur among
other tropidurids; hence, this character is unpo-
larized.

26. Trunk scales, lateral.—In Leiocephalus,
the lateral body scales are either equal in size to
(State 0), or appreciably smaller than (State 1)
those scales dorsal and ventral to them, as in L.
greenwayl, inaguae, macropus, melanochlorus,
psammodromus, and schreibersi. Small scales
co-occur with the presence of a well-developed
lateral fold (Character 34) in these species, ex-
cept for L. greenwayi. Dorsal scales that grade
into smaller ones laterally also occur in some
other tropidurids, for example “Stenocercus” and
Liolaemus. Therefore, this character is unpolar-
ized.

27. Middorsal crest.—A middorsal crest
(from an enlarged middorsal scale row) is present
in all but one species of Leiocephalus. However,
because of continuous variatton in size and shape
of the scales, the crest is difficult to characterize.
Only L. pratensis lacks a middorsal crest (State
2), whereas L. herminieri, lunatus,and personatus
possess a crest composed of prominent, attenu-
ate, overlapping scales (State 0). States interme-
diate to these extremes occur among the remain-
ing species (State 1). Although a middorsal crest
is found in most Tropidurinae, the crest may be
absent, well developed, or commonly, intermedi-
ate. Liolaeminae lack a middorsal scale row alto-
gether and, thus, no crest is present. This trans-
formation series is left unordered and unpolar-
ized.

28. Dorsal crest, scale number.—The mean
number of dorsal crest scales (occiput to vent) in
Leiocephalus varies from less than 50 in L. me-
lanochlorus to more than 65 in L. greenwayi,
inaguae, psammodromus, raviceps, and schrei-

20 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

bersi. The number of dorsal crest scales 1s contin-
uously variable (nondiscrete) in Tropidurinae,
ranging from few (<50) to many (>65). In Leio-
cephalus, the range of variation in the number of
dorsal crest scales occurs as two unpolarized
states—average number of dorsal crest scales
less than 60 (State 0) or 65 or more (State 1).

29. Postanal escutcheons.—A transverse row
of two, more often four, enlarged, dull white
escutcheon scales lies immediately posterior to
the vent in males of some species (State 1). So far
as I could ascertain, such scales are absent or rare
in tropidurids (State 0), but apparently do occur
in some eastern Tropidurus (D. Frost, pers.
comm.). In two taxa, Leiocephalus rhutidira and
vinculum endomychus, there are as many as 12
scales arranged in three or four rows (State 2).
None was observed in L. carinatus, greenwayi,
herminieri, melanochlorus, or punctatus; occa-
sionally, they are present in L. psammodromus.
Leiocephalus eremitus is known only from the
holotype female and cannot be scored.

30. Tricarinate toe scales (Fig. 13).—The
tricarinate scales at the base of the first and
second toes are enlarged into comblike fringes in
all Leiocephalus (State 0) (Etheridge, 1966a).
However, in some species (L. carinatus, eremitus,
pratensis), the comb is rather poorly differenti-
ated (State 1), whereas in L. barahonensis, and
some personatus (scalaris) and vinculum
(altavelensis), the comb is well developed and
prominent (State 2). Various tropidurids also
possess these enlarged tricarinate scales. I inter-
pret their simple presence in Leiocephalus to be
plesiomorphic, but whether the reduced comb is
a primitive step, or secondarily derived, is equivo-
cal. The transformation is unordered.

Fig. 13.
and second toe enlarged into combs (Leiocephalus
barahonensis, SOSNH 64582)(Character 30).

Tricarinate toe scales at the base of the first

SKIN FOLDS

31. Antebrachial folds.—Several skin folds
are common about the neck and body of iguanians.
Homologies of these structures are not well es-
tablished and their terminology has been applied
rather casually in the past. Frost (1987) attempted
to standardize the names of these folds and his
recommendations are followed here. Gular and
antehumeral folds are universally present in
Leiocephalus. Topographically, the gular fold is
like that of other tropidurids, but incomplete
medially and oriented more obliquely (\) than
vertically. Other neck folds are variously present
among Leiocephalus, such that three general con-
ditions obtain: simple—gular and_ short
antehumeral fold only; moderate—gular,
antehumeral, and oblique neck folds present; and
complex—gular, antehumeral, oblique neck,
longitudinal neck, and postauricular folds present.
Lateral neck folds are absent, or feebly present in
some “Stenocercus” and “Ophryoessoides” (e.g.,
O. iridescens, O. caducus, S. festae, S. aculeatus,
S. apurimicus), whereas in other tropidurids, neck
folds typically are complex. If one interprets
possession of complex folds as primitive, this
character is scored as a three-state transforma-
tion: complex (State 0) inL. eremitus, herminieri,
inaguae, macropus, melanochlorus, psammodro-
mus, and schreibersi; moderate (State 1) in L.
carinatus, cubensis, greenwayi, loxogrammus,
lunatus, personatus, punctatus, rhutidira, ravi-
ceps, semilineatus, stictigaster, and vinculum (v.
endomychus only); and simple (State 2) in L.
barahonensis, pratensis, and vinculum (except v.
endomychus). This character is unordered.

32. Antegular scale fringe.—Evident only in
some of the larger males of Leiocephalus
macropus and schreibersi (and 1 specimen of
personatus scalaris, USNM 224975) is a fold of
scales on the throat that is reminiscent topo-
graphically of the antegular fold of certain
tropidurids (e.g, Plica, Uranoscodon, various
Tropidurus). These are rather unlikely homologs,
however, because the structure in L. macropus
and schreibersiis aconvex demarcation of closely
spaced scales—i.e., a fringe, rather than an in-

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 21

tegumentary fold. The absence of the scale fringe
is primitive (State 0), but its presence (State 1) is
applied with hesitation because of such sporadic
occurrence in these species.

33. Nuchal fold.—A prominent nuchal fold is
evident in all species of Leiocephalus. The fold,
immediately posterior to the parietal scales, is a
straight, transverse cleft in L. barahonensis and
pratensis (State 1). The nuchal fold is moderately
convex in L. carinatus, cubensis, eremitus,
greenwayl, loxogrammus, lunatus, personatus,
punctatus, raviceps, rhutidira, semilineatus, stic-
tigaster, and vinculum (State 2), and is a strongly
convex, >-shaped fold in herminieri, inaguae,
macropus, melanochlorus, psammodromus, and
schreibersi (State 3). A nuchal fold is absent, or
at best, poorly developed in outgroup species
(State 0). I suspect that the transformation is
directional in Leiocephalus (from straight, to
moderate, to strongly convex), but that is infer-
ence, and not overly compelling; the character is
unordered.

34. Lateral fold——Some species of Leioce-
phalus possess a longitudinal fold on the side of
the trunk between the fore- and hind limbs. Ap-
parently, the presence of the structure is not a
function of large body size, nor is it homologous
with the dorsolateral fold of other tropidurids,
which is positioned more dorsally and is often
confluent with the antehumeral fold anteriorly
(Frost, 1987). A lateral fold like that of Leio-
cephalus is present in some Liolaemus, but is
otherwise absent (State 0) in tropidurids. Thus,
presence of a lateral fold probably is apomorphic
(State 1) for the following Leiocephalus: inaguae,
macropus, melanochlorus, psammodromus, ravi-
ceps, and schreibersi.

COLOR AND PATTERN

35. Ventral pattern.—As in outgroup taxa,
the venter of Leiocephalus is dull and nearly
patternless in some species, and boldly streaked,
spotted or smudged in others. Patterns often vary
subspecifically and only one ventral pattern could
be characterized: the presence of five to seven
complete transverse rows of single, dark con-
trasting scales in inaguae and schreibersi (State

1), and their absence elsewhere in Leiocephalus
and outgroup species (State 0).

36. Scapular patch.—Vivid, irregularly
shaped dark blotches above the forearm insertion
occur in male Leiocephalus lunatus, and in both
sexes of L. greenwayi and some macropus (e.g.,
macropus macropus). In the latter, the blotch is
bisected vertically by a thin white line, and in L.
greenwayi, there is also a dark patch above the
hindlimb. Such patches seem to be apomorphic,
because they are absent among outgroups, but
their homology is unclear, being sex-linked in
two species but not in the others. At the risk of
over simplification, the simple presence of a
scapular patch is scored as derived (State 1), and
its absence as primitive (State 0).

37. Suprascapular blotches.—In Leiocepha-
lus inaguae, melanochlorus, and psammodromus,
there are three or four dark, oblong blotches on
the suprascapular region. In L. melanochlorus,
the blotches continue onto the trunk and are vi-
vid, whereas in L. inaguae and psammodromus,
they fade and disappear. The blotches are scored
as derived in these three species (State 1). Their
absence, as in most outgroup species, is primitive
(State 0).

38. Facial band.—The presence of some form
of a broad, longitudinal band beginning from
behind the eye occurs in several species. In both
Leiocephalus cubensis and personatus, the band
is restricted to the side of the face as a mask. In
L. loxogrammus, macropus, pratensis, raviceps,
and semilineatus, the band continues onto the
neck and shoulder, and occasionally the trunk.
These markings are of dubious significance be-
cause similar features are not uncommon among
outgroup species. It is equally unclear as to
whether the restricted mask and band represent a
single transformation. The mask and band are
treated as separate states (1 and 2, respectively),
with State 0 being absence of a facial band or
mask.-The character is unpolarized and unor-
dered.

Throat patterns.—Among Leiocephalus,
throat patterns vary from nearly immaculate
(semilineatus) to faint streaks (carinatus), bold
chevrons (stictigaster), distinctive spotting
(Junatus, some personatus), or dull smudges

22 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

(barahonensis). Throat patterns often vary
subspecifically (e.g. cubensis, personatus) and
there is no meaningful characterization to be
obtained.

OTHER CHARACTERS

Hemipenis.—The hemipenis of Leiocephalus
is aunisulcate, weakly bilobate organ. Schwartz’s
(1967a:4) succinct description of the hemipenis
of L. lunatus is typical of all species: “The sulcus
is deep and prominent and is formed laterally by
an extensive membranous flap from the base of
the organ to near the tip. The non-sulcate surface
has a series of about four flounces (which extend
around the organ to near the sulcus) which rather
abruptly merge into a series of about six rows of
calyces. The tip of the hemipenis is smooth,
weakly bifurcate and much crenulated, the sulcus
extending into a cordate terminal area which
includes a very weak pair of papillae. From these
papillae, a raised area continues down the non-
sulcate surface, expands on its proximal half, and
ends at the level of the flounces on the non-
sulcate surface.” I found a modest amount of
variation among species in the number of basal
flounces, and in the degree of crenulations on the
tip of the hemipenis. This insignificant variation
precluded the hemipenis from further consider-
ation.

39. Tail cross section.—Primitively in Leioce-

phalus, the base of the tail is terete (State 0). In L.
herminieri, melanochlorus, psammodromus, and
inaguae, the base of the tail is laterally com-
pressed (State 1).

Karyology.—Karyotypes have been reported
for only a few species of Leiocephalus; each has
the presumably plesiomorphic iguanian number
of 12 metacentric macrochromosomes (Gorman
et al., 1967; Paul et al., 1976), Theres isesome
departure from the iguanian pattern of 24
microchromosomes, although the diploid comple-
ment of 12 + 24 occurs at least in L. schreibersi
(Gorman et al., 1967; Paul et al., 1976:17). Un-
published data of Hall (referred to in Paul et al.,
1976) mentioned “representatives of the Cuban
branch of the genus to have 12 + 20 patterns,” but
they failed to identify which species these were.
Recent work by Porter et al. (1989) disclosed a 12
+ 22 diploid number for L. carinatus, and a 12 +
18 number for L. raviceps, samples of both spe-
cies having been collected from the U.S. Naval
base at Guantanamo Bay. In the four male L.
raviceps, one microchromosome was much
smaller than the others, suggesting an XX/XY
sex chromosome system as in Uta and Sceloporus,
in which the minute microchromosome is pre-
sumed to be the Y-chromosome (Porter et al.,
1989). Obviously, karyological data for Leio-
cephalus are too incomplete at this time for use in
phylogenetic analysis; we only know that karyo-
types range from 2n = 30-36.

TREE TOPOLOGIES FOR LEIOCEPHALUS

The preceding 39 transformations (summa-
rized in Table 2 and Appendix I) were analyzed
with PAUP and HENNIG86. Two minimum-
length trees were found using PAUP, each having
118 steps and a Consistency Index (CI) of 0.441.
The two trees differ only in alternate relation-
ships of Leiocephalus personatus and lunatus
with respect to each other and to L. loxogrammus
plus raviceps. Tree no. 2 from the data output is
shown in Figure 14, the support for which is
discussed below. Alternate topologies (cf. Fig.
14) are also obtained using HENNIG86, although
again at 118 steps (CI = 44) as in the PAUP

analysis. Four of the alternate topologies pertain
to the melanochlorus group (see below); two
describe alternate relationships for personatus
and /unatus, with respect to Joxogrammus and
raviceps (the same two alternatives for Stem 4 in
Fig 14.), and there are three alternatives for the
placement of eremitus. Topologies for all other
terminal taxa do not vary with respect to Figure
14 except barahonensis.

In Figure 14 there are three clusters consisting
of (1) a clade of 11 species, mostly Hispaniolan
(L. barahonensis, lunatus, personatus, rhutidira,
semilineatus, vinculum, pratensis) but also three

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 23

Table 2. Summary of character-state transformations. U = unpolarized. UO = unordered.
1. Nasal overlap of premaxillary spine complete; spine not exposed dorsally posterior of external nares.
2. Premaxillary spine constricted basally.
3. Premaxillary spine with lateral projections above constriction.
4. Nasal-maxillary suture straight-sided anterolaterally.
5. Nasal-prefrontal contact reduced.(U).
6. Lateral wing and posterior process of septomaxilla reduced or absent.
7. Frontal bone broad and flat posteriorly.
8. Parietal table narrowly constricted posteriorly, (V-shaped).
9. Supratemporal bone lies ventomedially on supratemporal process of the parietal.
10. Proximal end of squamosal not expanded, distinct dorsal process present.
11. Rugosities well developed on skull roof.
12. Surangular notch well developed; angular process one-half or more the length of the surangular process.
13. Transition to tricuspid tooth crowns (dentary) at Tooth 3 or 4.(U)
14. Neural spines low, obtuse, and not expanded distally.
15 Centra of trunk vertebrae with narrow hypapophyses.
16. Frontonasal scale pattern Type I (State 1), Type I (State 2), Type II (State 3). (UO)
17. Median pair frontonasal scales enlarged.
18. Parietal scale pattern Type I (State 1), Type II (State 2), Type III (State 3). (UO)
19. Lateral postparietal scale enlarged.
20. Internasal scales three (State 1), two (State 2).
21. Five or six lorilabial scales anterior to elongate subocular.
22. Cephalic scales with ridges present but restricted (State 1) or ridges extend onto snout (State 2).

23. Enlarged temporal scale anterodorsal to ear.

WNNNN WN WV
SOO OoONNN Ff

. Two or three elongate temporal scales behind eye.

. Lateral neck scales small; keels reduced or absent. (U)

. Lateral trunk scales smaller than dorsal and ventral scales. (U)

. Dorsal crest present, moderate (State 1) or crest absent (State 2). (U, UO)

. Dorsal crest scales occiput-vent number 65 or more. (U)

. Postanal escutcheon scales 2—4 in single row (State 1); more than 4 in mulitple rows (State 2).
. Proximal scales of first and second toes enlarged into comb (2 steps). (UO)

31. Lateral neck folds moderate (State 1) or lateral neck folds simple (State 2). (UO)

Go
i)

. Antegular fringe present.

. Lateral fold present.

. Venter with 5—7 transverse rows of pale scales.
. Scapular patch present.

. Suprascapular blotches present.

WWW WW WwW WwW
Oo COA DA Nr fH W

. Tail laterally compressed.

Cuban (L. cubensis, stictigaster, raviceps), and
one Bahamian (L. loxogrammus); (2) a clade of
two western Hispaniolan species (L. melano-
chlorus, schreibersi), two southern Bahamian
forms (L. inaguae, psammodromus) and one
Cuban (L. macropus); and (3) a cluster composed
of two southern Bahamian species (L. greenwayi,
punctatus), the extinct L. eremitus, and the wide-

. Nuchal fold transverse (State 1), moderately convex (State 2), or strongly convex (State 3). (UO)

. Facial mask (State 1) and band (State 2) present. (U, UO)

spread L. carinatus. Leiocephalus herminieri is
the sister taxon of all other species.

Stems 9 and 10.—I have some confidence
that the 10 species above Stem 9 form a-natural
group. All taxa except Leiocephalus lunatus pos-
sess a dentary with an emarginate posterior bor-
der (12.1); with the exception of L. schreibersi,
this feature is found only in this clade. Skull

24 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

barahonensis

1 > cubensis
|__ §tictigaster
2 »-——loxogrammus

'__raviceps
___lunatus
____ personatus
a) ==rhutioira

___ semilineatus

vinculum

pratensis
lip tneguae
'_— pSammodromus

A _ melanochlorus

WA ml naCnOpUs
—— schreibersi
_______ carinatus
-—— greenwayi
punctatus

eremitus
herminieri

Fig. 14. Hypothesized relationships among species of Leiocephalus exclusive of fossil taxa based on 39
morphological characters (Appendix I) (PAUP, length = 118 steps, CI = 0.441). Support for the numbered stems is

given in the text and Appendix II.

rugosities (11.1) also are acquired at this stem,
but this character reverses in L. vinculum and
loxogrammus and is convergent in L. herminieri
and in greenwayi. Otherwise, Stem 9 is specified
by three other widely homoplastic characters—a
reversal to anarrow frontal (7.0), a ventromedially
placed supratemporal (9.1), and smooth cephalic
scales (22.0). Here, the peculiarities of L. pratensis
are evident. This taxon is not closely related to
any of the species above Stem 9 and is placed as
the sister species (Stem 10) by the common pos-
session of alternate states of four unordered,

unpolarized transformations, as follows. The pres-
ence of undifferentiated lateral neck scales (25.0)
occurs in L. carinatus, but is otherwise unique to
Stem 10. The snout scale pattern shifts from Type
II (16.2) to Type III (16.3), whence it reverses at
Stem 1 (cubensis + stictigaster). Antebrachial
folds shift from complex (31.0) to simple (31.2)
and then become moderate (31.1) at Stem 7. A
straight nuchal fold (33.1) changes to a moder-
ately convex nuchal fold (33.2) at Stem 8.
Stem 8.—The remaining nine species above
Leiocephalus pratensis and L. barahonensis at

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS

Stem 8 are uniquely united by a Type-II parietal
scale pattern (18.2). There is also a shift to a
moderately convex nuchal fold (33.2), which is
convergent at Stem 17. As a terminal taxon, L.
vinculum is restricted to the nominate subspecies
L. vy. vinculum from Ile de la Gonave. On the basis
of the present data and analysis, L. vinculum
altavelensis falls out with L. barahonensis,
whereas L. v. endomychus 1s regarded as a dis-
tinct species most closely allied with L. rhutidira.
The character support for this scheme and a
revised taxonomy are discussed in the Species
Accounts.

Stems 5—7.—Stem 7 unites four pairs of spe-
cies—cubensis + stictigaster and loxogrammus
+ raviceps ina branch with /unatus + personatus
and semilineatus + rhutidira (including endomy-
chus, which is not shown). The synapomorphies
of this clade (Stem 7) are presence of moderately
complex antebrachial folds (31.1) and a laterally
placed supratemporal process (9.0); the latter
represents a reversal (from Stem 9) to the
plesiomorphic condition for the clade, but shows
the derived condition in L. semilineatus. The
relationship between semilineatus and rhutidira
(Stem 6) is described by a straight nasal-maxil-
lary suture (14.1; convergent in /unatus) and the
presence of an enlarged temporal scale (23.1;
found elsewhere in barahonensis). Leiocephalus
semilineatus and L. rhutidira are linked with the
three other species pairs at Stem 5 only by one
widely homoplastic character—possession of
restricted cephalic scale ridges (22.1). An alter-
nate topology for Stem 7 is depicted in Figure 15
and discussed below.

Stems 2—4.—Two species pairs are united at
Stem 4 (Fig. 14; /unatus + personatus and loxo-
grammus + raviceps) by one uniquely derived
state—narrow hypapophyses on the trunk and
lumbar vertebrae (15.1). The presence of a large
lateral postparietal scale (19.1) also occurs at this
node, but this feature is reversed (19.0) in L. per-
sonatus and convergent in L. semilineatus. There
are no transformation states unique to L. per-
sonatus and L. lunatus (Stem 3); however, except
for L. herminieri, only these two taxa possess a
dorsal crest composed of attenuate, strongly over-
lapping scales (27.0), acharacter that was treated

NO
Nn

as unordered and unpolarized. An equally parsi-
monious alternative links L. /unatus as the sister
species of loxogrammus + raviceps based on
their common possession of a large lateral post-
parietal scale (19.1). This, of course, requires the
presence of attenuate, strongly overlapping dor-
sal crest scales (27.0) to be convergent in L.
personatus. The evidence uniting L. loxogrammus
and L. raviceps as sister taxa (Stem 2) is stronger;
only these two species have elongate
supratemporal scales behind the eye (24.1) and,
except for L. schreibersi, these are the only spe-
cies that possess low, obtuse neural processes
(14.1). Leitocephalus loxogrammus and L.
raviceps also display a facial band (38.2), but this
feature is widely convergent elsewhere on the
tree.

Stem 1.—Stem | (cubensis + stictigaster) is
specified by a shift from a Type-III to a Type-II
pattern of snout scales (16.2) and well-defined
cephalic scale ridges that extend onto the
frontonasals (16.2). Both of these transforma-
tions are homoplastic, but the sister relationship
between Leiocephalus cubensis and L. stictigaster
seems quite credible against the backdrop of the
entire tree. This analysis actually produces a
minimum stem length of 0 for stictigaster; how-
ever, this species 1s separable from L. cubensis by
the distinctly lineate dorsum and smaller body
size of L. stictigaster.

Stem 14.—The best supported branch on the
tree is that uniting species referred to herein as the
Leiocephalus melanochlorus group (inaguae,
psammodromus, melanochlorus, macropus,
schreibersi). Although there is only one apomorphy
unique to Stem 14 (a Type-I parietal scale pattern
[18.1]), three other transformations that occur at
this stem are specific save for usually single occur-
rences of the shared state elsewhere. These taxa
possess a basally constricted premaxillary spine
(2.1; also punctatus, and reversed in mela-
nochlorus), small lateral trunk scales (26.1; also in
greenwayi), and a distinct lateral fold on the trunk
(34.1; also in raviceps). Possession of a strongly
convex nuchal fold (33.1; also in herminieri) is
peculiar to these species as well, but as an unor-
dered transformation, this state first appears on the
ancestral stem of the tree.

26 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

——_ bDarahonensis
rhutidira

ae semilineatus

vinculum
ie cubensis

©)

___ Stictigaster
—— loxogrammus
6 \__ raviceps

7 ee nets

personatus
pratensis

18 (4

—— Inaguae

___ psammodromus
____ melanochlorus
macropus
schreibersi

WW

2

ees)

carinatus
Te 7

___ greenwayi
EE POUunCtarus

16

__ eremitus
herminieri

Bigs:

An alternate tree topology for Leiocephalus based on the same data set as Figure 14, but derived from

HENNIG86 (length = 118 steps, CI = .44). Support for numbered stems given in text.

Stems 11-13.— Within the Leiocephalus mela-
nochlorus group, L. macropus and schreibersi
are joined at Stem 13 by a Type-II snout scale
pattern (16.2; present also in L. inaguae and
several other species) and by the unique posses-
sion of the antegular scale row (32.1). At Stem 12
(melanochlorus, inaguae + psammodromus), the

only unique state is the acquisition of
suprascapular blotches (37.1). Additionally, how-
ever, the septomaxilla is reduced (6.1; except in
inaguae; convergent in /oxogrammus and
herminieri). The lorilabial scale row increases
(21.1; except in inaguae; convergent in eremitus
and punctatus). The squamosal acquires a dis-

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS A]

tinct supratemporal process (10.1; convergent in
macropus and some /oxogrammus) and the base
of the tail is laterally compressed (39.1; also in
herminieri). Stem 11 (inaguae and psammodro-
mus) iS denoted by the presence of a unique
lateral spikelike process above the basal con-
striction of the premaxillary spine (3.1), by in-
complete overlap of the nasals onto the premax-
illary spine (1.0), a V-shaped parietal table (8.1;
convergent in /oxogrammus, greenwayi, and
punctatus), the widely homoplastic states of
smooth cephalic scales (22.0) and an intermedi-
ate number of dorsal crest scales (28.1). Alternate
relationships within the L. melanochlorus group
are shown in Figure 16A—D and discussed below.

Stem 16.—Stem 16 describes a relationship
between Leiocephalus carinatus and the two
southern Bahamian taxa L. punctatus and
greenwayi. Of the three transformations at this
node one is unique—the possession of a con-
spicuously enlarged pair of median frontonasal

inaguae

psammodromus
3 melanochlorus

macropus

schreibersi

inaguae
psammodromus
melanochlorus
macropus

schreibersi

C

scales (17.1). The others are a shift from complex
to moderate antebrachial folds (31.1), and a
Type-II snout scale pattern. The relationship be-
tween greenwayi + punctatus (Stem 15) seems
odd phenetically because the small body scales
and distinct inguinal and scapular patches of L.
greenwayi contrast sharply with the larger body
scales and darker, unicolor pattern of L. punctatus.
Apart from L. carinatus, they posses a V-shaped
parietal roof (8.1; convergent with inaguae and
psammodromus) and the plesiomorphic, inter-
mediate state of tricarinate toe scales (30.0).
Leiocephalus greenwayi and L. punctatus con-
verge on the melanochlorus group in other ways,
but primarily in states that are interpreted as
primitive (e.g., incomplete overlap of the pre-
maxillary spine by the nasal bones [1.0]) or for
which the polarity is equivocal (e.g., reduced
nasal-prefrontal contact [5.1]). Additionally, L.
greenwayi possesses small body scales (i.e., a
high number of dorsal crest scales occiput-vent

inaguae
psammodromus
melanochlorus

macropus

schreibersi
inaguae
7 psammodromus
schreibersi
8 macropus

melanochlorus

D

Fig. 16. Alternate topologies for species of the Leiocepahlus melanochlorus group. A is also that of Figure 15

and B that of Figure 14.

28 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

[28.1]), a transformation that also characterizes
L. inaguae, psammodromus, schreibersi, and
raviceps, and small lateral trunk scales (26.1).
Similarly, L. punctatus has a basally constricted
premaxillary spine, as do most of the L.
melanochlorus group, and a frontal that is wide
and flat posteriorly, another feature of the L.
melanochlorus group, and one that also occurs in
L. pratensis.

Stem 17.—The inclusion of Lelocephalus
eremitus with L. carinatus, and greenwayi +
punctatus at Stem 17 is tenuous owing to the
absence of information on L. eremitus for the first
15 characters (osteological). Leiocephalus
eremitus falls out here on the basis of poorly
developed tricarinate toe scales (30.0; conver-
gent in pratensis and of ambiguous polarity) and
the intermediate state of a moderately convex
nuchal fold (33.2), which also appears at Stem 8.

Stem 18-19.—In spite of the aforementioned
similarities with the Leiocephalus melanochlorus
group, the Stem 17 clade bears no special rela-
tionship to them based on minimum step-trans-
formations. Rather, they configure at Stem 19 as
the sister group of all other Leiocephalus (except
L. herminieri) by possessing three internasal
scales (20.1). The L. melanochlorus group is
united at Stem 18 with all of the Stem 10 species
by five transformation states, all of which show
reversals at higher branches—viz., complete over-
lap of the nasals onto the premaxillary spine
(1.0); nasal-prefrontal contact that obscures the
nasal processes of the frontal (5.0); frontal bone
wide and flat posteriorly (7.1); cephalic scale
ridges present, but restricted (22.1), and the pres-
ence of postanal escutcheons (29.1). At this level
of the tree, only the presence of postanal escutch-
eons (29.1) that are absent in L. carinatus,
punctatus, and greenwayi (but also absent in
vinculum and melanochlorus) could be a synapo-
morphy.

Last, there is Leiocephalus herminieri, which
exhibits no evident affiliation with any other
species on the tree. It possesses a peculiar com-
bination of plesiomorphic and apomorphic states
that are scattered among other Leiocephalus (e.g.,
reduced septomaxilla, skull rugosities, Type-II
pattern of parietal scales, middorsal crest with
attenuate scales, strongly convex nuchal fold,

laterally compressed tail base). Only L. herminieri
possesses keeled ventral scales and retains a
large sternum. Also, it is the largest species of
Leiocephalus known from a whole specimen;
several fossil taxa were as large or larger, a fact
which engenders anxiety over possible ontoge-
netic influence on certain osteological charac-
ters, despite diligent attempts to correct for size
where appropriate. But L. herminieri is known
from only four specimens, one of which is a
skeleton and all of which are adults.

Comments and alternate hypotheses.—Pe-
rusal of the character-change list (Appendix III)
reveals a low consistency index for numerous
transformations, especially those that appear on
lower branches of the tree in Figure 14. Accord-
ingly, all the alternate topologies obtained with
HENNIG86, which, as mentioned above, affect
primarily Leiocephalus personatus, lunatus,
loxogrammus, raviceps, the melanochlorus group,
and eremitus, are plausible. One of the trees is
shown in Figure 15. Note that the major stems are
the same as in Figure 14. Further, L. vinculum is
in a group (Stem 3) with L. barahonensis paired
to rhutidira + semilineatus. All taxa except L.
rhutidira possess a ventromedially placed supra-
temporal (9.1; convergent in psammodromus)
and a reversal to smooth cephalic scales (22.0;
convergent in psammodromus and inaguae). Stem
2, uniting L. rhutidira and L. semilineatus plus
barahonensis, is specified by the enlarged tem-
poral scale anterodorsal to the ear (23.1). Al-
though unique to this branch, there is, as noted
previously, doubt about the homology of the
enlarged temporal scale and problems with its
variable occurrence in L. rhutidira and L. semi-
lineatus.

Four alternate topologies for the Leiocephalus
melanochlorus group (Stem 14) are shown in
Figure 16A—D. Each requires reversals or
convergences at more inclusive branches, but
also frequently with one or, occasionally, two
terminal taxa elsewhere, notably L. herminieri,
greenwayi, punctatus, and loxogrammus. On Stem
3 in Figure 16A, the distal ramus of the squamo-
sal narrows (10.1; also in some /oxogrammus),
the frontal widens and flattens posteriorly (7.1;
also in punctatus and pratensis), the internasals
are reduced to two (20.2; a widely homoplastic

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 29

character), and cephalic scale ridges extend onto
the frontonasals (21.2; also widely homoplastic).
Stem 4 (Fig. 16A) is equivalent to Stem 14 of
Figures 14 and 15. Stems 5—8 in Figure 16C—D
depict the distribution of equally parsimonious
alternate states to transformations used to define
topology 16A. None of these four topologies is
without problems, but topologies 16A and 16B
probably would be favored because they afford
less disturbance (homoplasy) to characters of
“confidence” on other stems.

Of the three options for the placement of

Leiocephalus eremitus, one is shown in Figure
14, the second in Figure 15, and the third is as the
sister taxon of all Leiocephalus exclusive of L.
carinatus, greenwayl, punctatus, and herminieri
(i.e., it would configure as the sister taxon of
Stem 18 in Fig. 14 and of Stem 15 in Fig. 15).
Each of these topologies requires broad ho-
moplasy, and probably none should be embraced
enthusiastically. Based purely on external mor-
phology because L. eremitus was scored only for
scales and folds, the suggested affinities of L.
eremitus lie with the L. melanochlorus group.

FOSSIL LEIOCEPHALUS

WEstT INDIES

Known fossils.—Six species of Leiocephalus
are known only by fossils from both the Greater
and Lesser Antilles. Few of the deposits from
which the bones were recovered are associated
with accurate chronologies based on radiocarbon
ages, but none is likely to be older than late
Pleistocene. Some are late Holocene age and
include species that became extinct at or near the
time of European settlement. Four of these spe-
cies are extralimital and document that in the
recent past Leiocephalus ranged throughout all
of the main islands of the Greater Antilles and
probably most of the islands in the Lesser Antilles
at least as far south as Martinique (Fig. 17).
Extralimital fossils in the Greater Antilles in-
clude one (possibly 2) species from Jamaica, L.
Jamaicensis (Etheridge, 1966b), and two from
Puerto Rico, L. etheridgeiand L. partitus (Pregill,
1981). Inthe Lesser Antilles, the extinct L. cuneus
(Etheridge, 1964) from Barbuda and Antigua
helps to fill an otherwise peculiar distributional
gap between the now extinct L. herminieri, pre-
sumed to have come from Martinique (cf. Spe-
cies Accounts), and the nearest living species on
Hispaniola. With fossils now known from other
Leeward and Windward islands (see below),
Leiocephalus evidently once ranged throughout
a good part of the Lesser Antilles.

The two fossil species that are known from
deposits located within the current range of
Leiocephalus are L. apertosulcus (Etheridge,
1965) and L. anonymous (Pregill, 1984), both

from Hispaniola (Fig. 17). These are the only
species of Leiocephalus that have an open
Meckel’s groove. In L. apertosulcus, the groove
is open and completely exposed medially
(Etheridge, 1965), whereas in L. anonymous,
Meckel’s groove is open except at the midpoint
of the dentary, at approximately the level of Teeth
12-15 where the groove is closed but not fused
(Pregill, 1984). In three other species—L.
etheridgei, herminieri, and partitus—Meckel’s
groove is open below at the level of Teeth 6-8,
whence the groove continues as a narrow sulcus
to the terminus of the jaw. THis condition reflects
incomplete fusion as well. Both L. apertosulcus
and L. anonymous have a well-developed intra-
mandibular septum (IMS). Except for L. cuneus
and L. partitus, the intramandibular septum is
greatly reduced or absent in all other Leiocephalus
and in other tropidurids in which Meckel’s groove
is closed and fused. Previously, I discussed the
ambiguous polarity of the open Meckel’s groove
in Leiocephalus and retention of the IMS in L.
cuneus and L. partitus and tentatively concluded
that both characters are derived reversals (Pregill,
1984). This still holds under the present scheme
with the Tropidurinae being the primary outgroup
and the Liolaeminae the second outgroup. The
open state occurs in several species of Liolaemus.

The extent of emargination at the posterior
end of the dentary (12) is difficult to determine
because the angular and surangular processes
often are incomplete or missing from fossils.

30 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

iS

ao
S inn

Xbaw

4] CENTRAL

Rigel.

atl FE
all HD PUERTO 5
=
t

{gue

The West Indies showing the present range (shaded) of Leiocephalus and those islands from which the

six paleospecies (F) are known (Jamaica, jamaicensis; Hispaniola, anonymous, apertosulcus; Puerto Rico,
etheridgei, partitus; Lesser Antilles, cuneus). Two other species (eremitus from Navassa Island, and herminieri
presumably from Martinique) have become extinct historically (7).

Two species, L. etheridgei and L. partitus, appear
to possess the derived condition (12.1).

None of the fossil taxa exhibits the extreme
anterior transition to tricuspid teeth (13.1) as in
Leiocephalus rhutidira and L. endomychus, but a
transition complete by Tooth 4 or 5 in L. cuneus
and L. etheridgei is within allowable limits of the
derived state of that character.

Several fossil species are also represented by
cranial elements other than dentaries that can be
scored according to the transformations (1—13) in
the primary data matrix (Appendix I); these are
summarized in Table 3. For three of these trans-
formations, it would be hazardous to estimate
character states because articulated skulls are
needed for an accurate interpretation. These trans-
formations are, first, the completeness of nasal
overlap of the premaxillary spine (1), second, the
straight-sided nasal-maxillary suture (4), and,
third, the position of the supratemporal (9). In

fossil taxa in which the premaxilla is known
(anonymous, Cuneus, and jamaicensis), there is
indication of neither a constriction at the base of
the nasal process (2.1) nor small lateral processes
above the constriction (3.1).

Of the five species that have referred frontals,
Leiocephalus anonymous, L. apertosulcus, and
L. cuneus show evidence that the nasal process of
the frontal remained exposed between the pre-
frontals and nasals (5.1), a transformation that
was not polarized. In L. anonymous, etheridgei,
and jamaicensis (1 specimen), the frontals are
wide and flat posteriorly (7.1) and in L. anony-
mous, apertosulcus, cuneus, and jamaicensis (1
specimen), rugose dermal sculpture is present
(11.1). The two frontals referred to L. jamaicensis
differ in their proportions and degrees of dermal
sculpture; possibly, this indicates that more than
one species is represented (cf. Species Accounts).
None of the fossil frontals has the extensive

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 31

Table:

Some osteological character states for fossil species of Leiocephalus. Transformations 1—13 are

those from the primary data matrix (text and Appendix I). MG = Meckel’s groove (open, open anteriorly, or
fused); IMS = intramandibular septum present. A dash (—) indicates that there is no corresponding bone referred,
or that the transformation is unascertainable. ? = species assignment questionable for that bone (see text).

Species 1 2 3 4 5) 6
anonymous = 0) 0 = =
apertosulcus - — = = =
cuneus — (0) 0 = ts

etheridgei ~
Jamaicensis —
partitus — — _ = = —

S|

S|

Ie al

Co OR ee
|

sculpture seen in such species as L. barahonensis
and L. personatus, which suggests that there
might be more than two states to this character.

Of the four species with referred parietals
(Table 3), Leiocephalus apertosulcus and L.
cuneus have the strongly V-shaped condition (8.1).

The impact of the addition of the six fossil taxa
to the primary data matrix (Appendix 1) on the
tree topologies of Figures 14 and 15 is the cre-
ation of polytomies at or near terminal branches.
The cladogram in Figure 18 is the result of both
STRICT and ADAMS Consensus Trees after
adding the taxa and data from Table 3. All basal
branches of Figures 14 and 15 are preserved. The
fossil species (denoted by an asterisk) are clus-
tered among various terminal branches owing to
the possession of one or two synapomorphies of
amore inclusive group. For example, L. cuneus,
L. apertosulcus, and L. anonymous constitute the
sister group of L. greenwayi, whereas L. etheridgei
and L. partitus are depicted as sister species of L.
schreibersi in the L. melanochlorus group.
Leiocephalus jamaicensis falls out as the sister
taxon of the species above Stem 10 in Figures 14
and 15. With so few applicable data available
from the fossils, I am indifferent toward the
relationships depicted in Figure 18.

New fossils from the Lesser Antilles.—Fos-
sils collected recently from Anguilla and
Guadeloupe represent new island records for
Leiocephalus. They are described below, and are
referred to:

Character
ql 8 9 10 11 1D 13 MG IMS
| 0) = = 1 0 0 D 1
0) 1 = = | 0) 0 p} |
0) l = = l 0 i 0 |
1 = = = 0) l 1 | 0
1? 0) = = Li, 0 0 0 0)
~ — _ = 1 0) 1 l

Leiocephalus cf. L. cuneus

Anguilla: Center Cave. One partial left dentary,
one partial frontal (SDSNH uncat.), by Gregory
Pregill, David Steadman, Ronald Crombie, and
Linda Gordon, 12 October 1982. Latest Holo-
cene.

Guadeloupe: Grande Terre, unnamed cave at
Pnte. du Capucin, ca. 2 km. S Pnte. de la Grande
Vigie. One partial left dentary, one maxillary
fragment, one sacral vertebra (SDSNH uncat.),
by Gregory Pregill, Richard Thomas, and Frank
Davis, 14 March 1984. Latest Holocene.

Description: The dentary from Center Cave,
Anguilla, is missing its posterior aspect near the
level of the last tooth. What remains is 8 mm
long. Meckel’s groove is open anteriorly as a
narrow slit below the most anterior four teeth. An
intramandibular septum is present. Most of the
teeth remain and all but the four anterior ones
have blunt but evidently flared tricuspid crowns.
The bone came from an individual with an esti-
mated snout vent length of 65—70 mm.

The dentary from Guadeloupe is missing the
anterior tip, but is otherwise complete. With a
tooth row of about 12 mm, the bone came from an
individual estimated at 110 mm, SVL. The labial
surface bears a wedge-shaped scar posteriorly
that marks the position of the coronoid overlap in
life and extends anteriorly below the most poste-
rior two teeth. There are 19 teeth or alveoli,
accounting for the most anterior one or two that

UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

—_— barahonensis
__ rhutidira
_— semilineatus
vinculum
=== CUBeNSIS
|____ stictigaster
a loxogrammus
meee

_—— l/unatus
-———_ personaitus
Jamaicensis*
pratensis
inaguae
psammodromus
melanochlorus
macropus
schreibersi
|___ etheridgei*
a partitus*
carinatus
greenway!

| ee anonymous*

| = apertosulcus*
| sss al cuneus*
| ee ees UME NAULLS
he ae eremitus
| - ; herminieri

Fig. 18. Consensus tree (STRICT and ADAMS) for Leiocephalus with the six fossil species (denoted
by asterisk, data from Table 3) added to the primary data matrix (Appendix 1). CF (normalized) = 0.923, Cl
= ().484. See text and Figures 14 and 15.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 33

accompanied the broken tip. The transition from
simple, pointed crowns to tricuspid crowns is
complete at the fifth tooth. The bone is worn and
abraded, and incised with what appear to be tooth
marks from a small rodent.

Because they are so fragmentary, none of the
other three bones requires much comment. The
frontal from Center Cave is 8.4 mm long
midsagittaly, neither wide nor flat posteriorly,
and came from an individual with an estimated
SVL of 60 mm.

Of the two dentaries, the Guadeloupe speci-
men is the more nearly complete. Otherwise they
differ from one another only in size. Both are
small compared with the type series of
Leiocephalus cuneus from Barbuda (Etheridge,
1964) and the material known from Antigua
(Pregill et al., 1988). They are referred to L.
cuneus on the basis of having (1) an intraman-
dibular septum, (2) a Meckel’s groove that is
closed and fused except for a shallow sulcus at
the anterior end of the jaw, and (3) the transition
to tricuspid teeth complete at the fifth tooth.
Because L. cuneus may have reached a snout-
vent length approaching 200 mm, the Anguilla
and Guadeloupe specimens (if correctly assigned)
must have represented juveniles or subadults.
Additional dentaries and other cranial elements
would be needed to confirm the identity of the
extinct populations on these two islands. Their
assignment here is necessarily tentative.

Both the Anguilla and Guadeloupe popula-
tions evidently persisted into historical times.
The Center Cave material is associated with a
radiocarbon age, based on charcoal, of 730 + 60
yr B.P (Pregill and Steadman, MS). There are no
associated dates with the fossils from Pnte. du
Capucin, but bones of Rattus (a European intro-
duction) were among the remains, all of which
were collected from the surface of a shallow
ledge.

NorTH AMERICAN TERTIARY

Several fossils from Oligocene and Miocene
deposits of North America have been compared
with, referred to, or identified as Leiocephalus.
Most are jaw fragments that have been referred to

the genus on the basis of their having flared
tricuspid tooth crowns, a fused Meckel’s groove,
and, in more nearly complete specimens, an indi-
cation of a coronoid labial blade on the dentary.
If these fossils were correctly assigned, the pres-
ence of Leiocephalus in the middle Tertiary of
North America would be a curious zoogeographic
puzzle indeed, as well as a potentially valuable
source of character-state information. However,
none of the fossils is diagnostic of Leiocephalus,
and in fact one referred series probably is better
placed with another iguanian taxon.

The oldest of these fossils is a jaw fragment
(CM 33650) from the late Oligocene Cedar Ridge
Local Fauna of central Wyoming (Setoguchi,
1978:15). This bone is merely a scrap bearing
four teeth with unflared crowns. It is far too
meager for generic assignment of any sort,
Setoguchi’s (1978) reference to Leiocephalus
being based on Estes’ (1963) characterization of
fossil dentaries from the early Miocene Thomas
Farm Locality of Gilchrist County, Florida. The
fossils from Thomas Farm consist of two partial
dentaries, several jaw fragments, and tentatively
associated vertebrae and girdle elements. Char-
acters suggesting Leiocephalus are a closed
Meckel’s groove, a scar marking the overlap of
the coronoid labial blade, tricuspid tooth crowns,
and pitting at the tooth bases. Referral to
Leiocephalus was a matter ‘of eliminating other
pleurodont iguanians with an open Meckel’s
groove, those that lacked a coronoid labial blade,
or that differed in details of tooth morphology.
However, there is nothing diagnostic about them,
and being so incomplete their identification nec-
essarily must remain inconclusive.

Robinson and Van Devender (1973) question-
ably identified as Leiocephalus a single anterior
half of a right dentary from the Monroe Creek
Formation, early Miocene of Nebraska. The speci-
men has a closed Meckel’s groove (except for a
large alveolar foramen anteriorly) and a single
tricuspid tooth that, based on their illustration
(Robinson and Van Devender, 1973:fig. 1D), is
only weakly flared at the crown. They compared
the fossil in detail with numerous iguanian: liz-
ards, but there is nothing about the bone that
unambiguously identifies it as Leiocephalus.

From the Norden Bridge Quarry, Mio-Plio-

34 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

cene Valentine Formation, Nebraska, Estes and
Tihen (1964:fig. SA—B) described a fragment of
a right dentary as “Unidentified iguanid, form
B.” On the basis of the four teeth with narrow
shafts and smoothly flared tricuspid crowns, they
concluded that the fossil“... appears to resemble
the Recent West Indian iguanid Leiocephalus in
the general proportions of the teeth and simple
flare of the crown (1.e., without thickening of the
shaft below the side cusps), but is too incomplete
to be identified with any confidence” (Estes and
Tihen, 1964:466). I concur.

The largest series of the Tertiary fossils was
collected at Annies Geese Cross Quarry in the
Lower Valentine Formation of eastern Nebraska
(Wellstead, 1982); additionally, there is one left
dentary from nearby Railway Quarry B (Holman
and Sullivan, 1981). Wellstead (1982) described
the numerous partial dentaries and maxillary frag-
ments as a new species, Leiocephalus septen-
trionalis. Because of homonymy with L. sticti-
gaster septentrionalis Garrido (1975), the name
was emended to L. nebraskensis (Wellstead,
1983). Wellstead (1982) also referred to this spe-
cies the single specimen from the Norden Bridge
Quarry described by Estes and Tihen (1964).

The holotype (UNSM 56085) is a nearly com-
plete left dentary with smoothly flared tricuspid
teeth, aclosed Meckel’s groove, and a scar on the
posterolabial surface marking the position of the
coronoid labial blade in life. Leiocephalus
nebraskensis was thought to differ from other
Leiocephalus by its possession of a faintly devel-
oped subdental ridge posteriorly and its more
robust size. My reexamination of the type series
revealed that, aside from shape of the tooth
crowns, L. nebraskensis lacks critical features of
not only Leiocephalus, but tropidurids generally.
A paratype right dentary (UNSM 47148) is re-
vealing. The dentary-postdentary articulation is
entire and the coronoid, the splenial, the anterior
half of the angular, and part of the surangular are
present. The dentary-postdentary articulation dif-
fers from that of Leiocephalus in several ways.
(1) The posterior end of the dentary terminates
below the level of the coronoid apex, rather than
extending posteriorly over the surangular. (2)
The angular is much larger than in Leiocephalus
and (3) the splenial is large, trapezoidal, and

terminates posteriorly near the level of the
coronoid apex. The splenial of Leiocephalus and
the Tropidurinae is straplike and extends posteri-
orly beyond the apex of the coronoid. Some of
these differences in the dentary-postdentary ar-
ticulation also can be deduced from the holotype.
The bone is noticeably expanded posterolingually,
implying that a large splenial was present and
that there was a substantial investiture of the
surangular in the dentary-postdentary articula-
tion. The lower jaw of L. nebraskensis is actually
more characteristic of the iguanid type of archi-
tecture described above. Except for tooth crown
morphology the fossils could be comfortably
assigned to Dipsosaurus dorsalis, a conclusion
also reached by Norell (1989).

The teeth of Leiocephalus nebraskensis have
smooth, fleurs-de-lis crowns similar to those of
most Leiocephalus (and some corytophanids,
Lang, 1989). The posterior teeth of Dispsosaurus
dorsalis have broader crowns, are more laterally
compressed, and usually have four cusps. Tricus-
pid tooth crowns are primitive for iguanines,
whereas presence of four or more cusps is thought
to be successively derived (de Queiroz, 1987).
The posterior teeth of iguanian lizards are highly
variable. Among Leiocephalus for instance, they
may have quite narrow, weakly tricuspid crowns
(e.g., L. partitus, Pregill, 1981) or may be nearly
molariform as in some L. psammodromus. In
light of this variability, there 1s no reason to
attach special significance to the tricuspid teeth
of these Tertiary fossils. For L. nebraskensis, the
nature of the dentary-postdentary articulation is
sufficient to justify removal of this species from
Leiocephalus.

To summarize, several partial or nearly com-
plete lizard dentaries from middle Tertiary de-
posits of North America have been referred to, or
identified as, Leiocephalus on the basis of the
possession of flared, tricuspid tooth crowns, a
closed Meckel’s groove, and a coronoid labial
blade. These features are characteristic of
Leiocephalus, but not unique. In most cases, the
referred specimens are too incomplete to assign
them confidently to any iguanian taxon. In the
case of Leiocephalus nebraskensis, 1t would be
more appropriate to refer the taxon to a non-tropi-
durid genus, perhaps Dipsosaurus.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS

Ww
Nn

BIOGEOGRAPHY

Leiocephalus is monophyletic and endemic to
the West Indies, having been distributed through-
out most of the major islands or banks until
recently. As a tropidurid iguanian, its ancestry 1s
tied to South America. The radiation of
Leiocephalus in the West Indies was not explo-
sive like that of Anolis, but neither was it insig-
nificant for a terrestrial squamate. The distribu-
tion and numbers of species of Leiocephalus are
comparable to, or greater than, those of other
ground-dwelling lizards. For example, diplo-
glossines have about 20 Antillean species, but
they do not occur south of Montserrat. Ameiva
has 18 species ranging throughout the Greater
and Lesser Antilles and the Bahamas; however,
there is no strong evidence that they constitute a
natural group apart from mainland Ameiva (but
see Barbour and Noble, 1915; Baskin and Wil-
liams, 1966). Leiocephalus is, of course, the only
spiny, scansorial iguanian in the West Indies and
most species are xerophilous. The greatest diver-
sity of Leiocephalus (12 species) is found on
Hispaniola; this is the case for most all other
terrestrial reptiles in the Antilles and doubtless is
related to the island’s size and its habitat diver-
sity. Cuba has six species (with one being extra-
limital) and the Bahamas five (with all but L.
carinatus being outside the Great Bank). Puerto
Rico has two extinct species, Jamaica one, or
possibly two, extinct species, and two, or possi-
bly three, species are known from all of the
Lesser Antilles. Only two of the six fossil species
are from an island where Leiocephalus still ex-
ists; all were as large or larger than any extant
species.

The earliest West Indian radiation could have
begun on any of several islands or banks, whether
ancestral Leiocephalus originated from over-
water dispersal or as a faunal component of a
proto-Antillean block. If Leiocephalus herminieri
is the sister species of all other Leiocephalus, its
presumed distribution on Martinique suggests a
southern entry for the genus into the Antilles, or
at least an early presence there. However, L.
herminieri is not especially plesiomorphic; thus,

one cannot argue persuasively for an application
of Hennig’s (1966) Progression Rule of character
transformation in peripheral isolates (see also
Wiley, 1981), especially in the absence of recov-
erable character information on the extinct spe-
cies of Leiocephalus to the north. Nevertheless,
L. herminieri is important for understanding the
historical distribution of Leiocephalus, with or
without reference to a center of origin. Similarly,
Leiocephalus carinatus is in many respects the
most generalized and least apomorphic species,
and also has the broadest range. It is the only
species not endemic to a single island (or bank),
being common throughout Cuba, Grand Cay-
man, Little Cayman, Cayman Brac, and the Great
and Little Bahama Banks (Fig. 20).

Indeed, not much is to be gained by hypoth-
esizing an initial center of radiation of
Leiocephalus in the West Indies, be it Cuba,
Hispaniola, the Bahamas, or the Lesser Antilles,
because the basal radiation of Leiocephalus may
be older than the Antilles themselves. If that
radiation is not older than the Antilles and if the
phylogeny proposed here approximates reality,
then Cuba, Hispaniola and the Bahamas have
experienced multiple invasions by Leiocephalus.
The L. melanochlorus group, for example, is
composed of two southern Bahamian taxa
(psammodromus and inaguae), two Hispaniolan
(melanochlorus and schreibersi) and one Cuban
(macropus). Another Cuban species, L. raviceps,
is most closely allied to the Bahamian species L.
loxogrammus and, together, they belong to an
Hispaniolan subgroup (personatus and lunatus)
which inclusively has two additional Cuban mem-
bers, L. cubensis and L. stictigaster. This analysis
found no strong evidence of a Leiocephalus
personatus-complex (personatus, lunatus,
barahonensis, vinculum) in the sense of Cochran
(1941) or Schwartz (1967a), but there is some
suggestion of other species clusters on Hispaniola
(e.g, rhutidira + endomychus + semilineatus).

The distributional patterns of Leiocephalus
are consistent with the fact that the species com-
pose a group of active, diurnal lizards that evolved

36 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

in an archipelago with a complex history of geol-
ogy and climate (Pregill and Olson, 1981; Wil-
liams, 1989). Additionally, there is a compara-
tively low level of differentiation among spe-
cies—i.e., a lack of many autapomorphies. One
might be persuaded that Leiocephalus has not
been in the Antilles all that long, if differentiation
is a function of time. But such conjecture is
hazardous without knowledge of evolutionary
rates. Moreover, it could be argued that the habitus
of Leiocephalus is so like that of all other small,
scansorial iguanians (e.g., numerous phrynoso-
matids and tropidurids) that it must have evolved
long ago as a fundamental adaptation, and as an
historical constraint, that is the reason for so
much similarity among the species.
Leiocephalus has been more vulnerable to
extinction and extirpation during the Holocene
than any other West Indian squamate, with the
possible exception of colubrid snakes (Alsophis

spp.). Several factors are probably at play. Else-
where I suggested that the timing of their extinc-
tions is strongly correlated with human settle-
ment of the islands (Pregill, 1986). Most other
squamates have managed to cope at some level
with man and his commensals, so why has
Leiocephalus been so much more prone to ex-
tinction? The little we know about the natural
history of these lizards suggests that the cause is
not a peculiarly restricted diet, because they
readily consume a variety of arthropods and,
opportunistically, will feed on most anything
including plants and other lizards (Schoener et
al., 1982; Armas, 1987; pers. obs.). It is possible
that some of the species have a relatively low
(compared with Anolis, for instance) reproduc-
tive potential and/or delayed sexual maturity. If
so, increased predation by alien species would
take its toll, especially in smaller habitat areas.

SPECIES ACCOUNTS

Twenty-nine species of Leiocephalus are in-
cluded in the following accounts. Six of these are
fossil species and all are West Indian. One extant
species, L. endomychus, is proposed as a new
combination, and L. vinculum altavelensis is re-
assigned as L. barahonensis altavelensis. The
accounts are presented alphabetically and in-
clude authors, synonymies, holotype, and gen-
eral statements of distribution (see also Figs. 19—
21). For purposes of description, a characteriza-
tion is provided in place of a strict diagnosis
(autapomorphies only). Known subspecies and
their authors are included where germane.

LEIOCEPHALUS GRAY (1827)

Holotropis Duméril and Bibron, 1837 (part).
Pristinotus Gravenhorst, 1837.
Steironotus Fitzinger, 1843.
Hispaniolus Cochran, 1928a.
Leiocephalus Etheridge, 1966a.
Type species.
(1827)
Content.—Twenty-nine species (21 extant; 2
extinct historically; 6 fossil species of late Qua-
ternary age).

Leiocephalus carinatus Gray

Distribution.—West Indies. Now restricted
to Cuba, Hispaniola, and the Bahamas. Formerly
ranged throughout the Greater Antilles, and Lesser
Antilles south to Martinique (Figs. 19-21).

Characterization.—Terrestrial iguanian
squamates of small (53 mm SVL) to large (to
200 mm SVL) size; nasal bones large; septo-
maxilla reduced and free posteriorly; parietal
roof constricted posteriorly in adults; nasal proc-
ess of premaxilla overlapped dorsally and ven-
trally by nasal bones; Meckel’s groove usually
closed and fused; tooth crowns variable, but
always tricuspid posteriorly; anterior process
present on interclavicle; posterior process of
interclavicle broadly flared; sternum reduced
(except L. herminieri); xiphisternal rods curve
anteriorly crossing over postxiphisternal ribs;
ribs on cervical vertebrae expanded as scoops;
caudal neural spine sail-shaped; autonomic frac-
ture planes beginning on fifth or sixth caudal
vertebra; cephalic scales large, platelike; en-
larged subocular scale present; body scales
keeled; ventral scales smooth (except L.
herminieri); antebrachial folds simple to com-
plex; tail-curling behavior common.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS Sif

150 miles

ars San Salvador

S.: ras os Rum Cay

ae Silt Brac oar

~ Grand Cayman

Figs (9!
Bahamas, and Hispaniola. (See also Figs 17 and 20).

Leiocephalus anonymous Pregill, 1984

Holotype.-—USNM(VP) 340182. Right
dentary. Type locality: an unspecified cave(s)
near St. Michael de |’ Atalye, Dépt. L’ Artibonite,
Haiti. Fossils collected by Arthur J. Poole (ca.
1927) in cave sediment probably no older than
latest Pleistocene.

Distribution.—Extinct, known only by fos-
sils from the type locality in Haiti.

Characterization.—Large size (to 130 mm
SVL); Meckel’s groove open except for a short
distance in the middle of the dentary, usually
between Teeth 12—15 where the upper and lower
borders converge and touch; open portions of
Meckel’s canal exposing a well-developed
intramandibular septum; parietal foramen located
wholly within frontal bone.

4 é Samana Cay
:: Crooked

Bod es. Plana Cays

_ Pacsins > Maou

Great
Inagua” 45

Gonave ' <

Principle islands of the West Indies north of Puerto Rico. Extant Leiocephalus occur only in Cuba, the

Remarks.—Pregill (1984) referred numerous
dentaries and other cranial elements to this spe-
cies. He speculated that L. anonymous may have
persisted into historical times.

Leiocephalus apertosulcus Etheridge, 1965

Holotype.-—MCZ(VP) 3404. Right dentary.
Type locality: Stratum 2, cave in Cerro de San
Francisco, Municipio Pedro Santana, San Rafael
(=La_ Estrelleta) Province, Republica
Dominicana.

Distribution.—Extinct, known only by fos-
sils from the type locality in the Dominican
Republic.

Characterization.—Large size (to 150-200
mm SVL); Meckel’s groove completely open and
exposed on the lingual side of the dentary.

38 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

uo carinatus

carinatus

aie

loxogrammus

Fig. 20. Distribution of Bahamian species of Leiocephalus. Islands of the Great and Little Bahama Banks are
enclosed at 100 fathoms. See Figure 19 for names of islands.

Remarks.—Etheridge (1965) referred numer-
ous cranial and postcranial bones to this species;
its extinction was presumed to be pre-Columbian.

Leiocephalus barahonensis Schmidt, 1921a

L. altavelensis Noble and Hassler, 1933.
L. personatus barahonensis —Mertens, 1939;

Cochran, 1941.

L. vinculum altavelensis —Schwartz, 1967a.
L. barahonensis —Schwartz, 1967a.

Holotype-—AMNH 2736. Type locality:
Barahona, Barahona Province, Reptblica
Dominicana.

Distribution.—Southcentral Hispaniola, pri-
marily Barahona Peninsula, and west to Jacmel,
Dépt. du Sud-Est, Haiti; Isla Beata, Isla Alto
Velo.

Characterization.—Moderate size (males to
80 mm, females to 65 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary
suture curved; nasal processes of frontal not ex-
posed; frontal narrow; parietal table U-shaped in
adults; supratemporal ventromedial; supratem-
poral process of squamosal indistinct; skull roof
rugose; angular process of dentary well devel-
oped; transition to tricuspid crowns at Tooth 10
or 11 on dentary; snout scale pattern Type HI (2
rows of scales between internasals and anterior
pair of frontals, with posterior row composed of
3, or a single pair of, enlarged scales); parietal
scale pattern Type III (most lateral parietals larger
than median pair, postparietals few or absent); 3
internasal scales, usually 2 in contact with ros-
tral; 4 lorilabial scales anterior to enlarged
subocular; head scales smooth; single, enlarged

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 39

ee

Anguillay
St. Martin :

Dominica’.

.
<ie

Fig. 21. Islands and banks of the Lesser Antilles. Leiocephalus herminieri is thought to have come from
Martinique, where it is now extinct. Fossils of Leiocephalus are known from (north to south) Anguilla (cf. cumeus),

Barbuda/Antigua (cuneus), and Guadeloupe (cf. cuneus). The material from Anguilla and Guadeloupe is new and
is described in the text.

40 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

temporal scale; lateral neck scales keeled, undif-
ferentiated; lateral trunk scales not reduced; mid-
dorsal crest moderate, 42-57 dorsal crest scales
occiput—vent; usually 4 postanal escutcheons;
tricarinate scales of first and second toe enlarged,
forming prominent fringe; lateral neck folds
simple; nuchal fold transverse; lateral fold on
trunk absent; distinctive scapular or suprascapular
pattern absent.

Subspecies.—barahonensis Schmidt (192 1a);
altavelensis new combination (see Remarks);
aureus Cochran (1934b); beatanus Noble (1923);
oxygaster Schwartz (1967a).

Remarks.—Schwartz (1967a) reviewed the
subspecies of Leiocephalus barahonensis and
provided details on scutellation and color pattern
based on numerous specimens. He puzzled over
the affiliation of altavelensis, a population of
Leiocephalus isolated on Isla Alto Velo off the
southern tip of the Barahona peninsula, and sug-
gested that the population might represent (1) an
aberrant form of barahonensis, (2) a separate
species as originally described by Noble and
Hassler (1933), or (3) a subspecies of vinculum,
despite its peculiar distribution far removed from
the nearest putative vinculum on Ile de la Gonave.
Of these three options, Schwartz (1967a) least
preferred the first (as proposed here) because of
the smaller size of altavelensis and what he re-
garded as distinct scale and chromatic differ-
ences. However, L. altavelensis shares with L.
barahonensis conspicuously enlarged tricarinate
toe scales (30); a transverse nuchal fold (33); and
a Type III parietal scale pattern (17); in addition,
L. altavelensis has a moderately enlarged tempo-
ral scale (19). Although the polarity of each of
these characters is equivocal, none of these states
occurs in nominate vinculum. Eventually, L.
altavelensis may prove to be distinct from
barahonensis, but the available evidence favors
the interpretation proposed here.

Leiocephalus carinatus Gray, 1827

Holotropis microlophus Duméril and Bibron,
1837 (part); de la Sagra, 1837.

Leiocephalus macleayti Gray, 1845.
Holotype-—BMNH 1946.8.29.75. Type lo-

cality: restricted by Schwartz and Ogren (1956)
to La Habana, Habana Province, Cuba.

Distribution.—Cuba, island-wide and essen-
tially coastal; Isla de la Juventud; Cayman Is-
lands, Great Bahama Bank, Little Bahama Bank;
introduced in south Florida and on Great Swan
Island.

Characterization.—Large size (males to 130
mm, females to 110 mm SVL); nasal overlap of
premaxillary spine usually complete; nasal-max-
illary suture curved; nasal processes of frontal
exposed in some; frontal narrow; parietal table
U-shaped in adults; supratemporal lateral;
supratemporal process of squamosal indistinct;
skull roof usually smooth; angular process of
dentary not well developed; transition to tricus-
pid crowns at Tooth 10 or 11 on dentary; neural
spines of trunk vertebrae obtuse and not ex-
panded distally in some; snout scale pattern Type
II (3 rows between internasals and anterior
frontals, with posterior row composed of 3 scales
in contact with anterior frontals); enlarged pair of
frontonasals; parietal scale pattern Type III (most
lateral parietals larger than median pair,
postparietals few or absent); 3 internasals, usu-
ally all in contact with rostral; 4 lorilabial scales
anterior to enlarged subocular; cephalic scales
smooth; lateral neck scales keeled, undifferenti-
ated; middorsal crest moderate, 43-60 dorsal
crest scales occiput—vent; postanal escutcheons
absent; tricarinate scales of first and second toe
not greatly enlarged, fringe weakly developed;
lateral neck folds moderate; nuchal fold moder-
ately convex; lateral fold on trunk absent; dis-
tinctive scapular or suprascapular pattern absent.

Subspecies.—carinatus Gray; aquarius
Schwartz and Ogren (1956); armouri Barbour
and Shreve (1935); cayensis Schwartz (1959a);
coryi Schmidt (1936); granti Rabb (1957);
hodsoni Schmidt (1936); labrossytus Schwartz
(1959a); microcyon Schwartz (1959a); mogo-
tensis Schwartz (1959a); varius Garman (1887):
virescens Stejneger (1901); zayasi Schwartz
(1959a).

Remarks.—Although I examined Leio-
cephalus carinatus from most populations
throughout its range, the characterization is based
primarily on specimens of L. c. aquarius col-

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 4]

lected from the U.S. Naval Base at Guantanamo
Bay. However, Bahamian L. carinatus are con-
sistent with respect to these transformations. Some
clinal variation in snout-vent length and several
meristic features has been demonstrated (Rabb,
1957). Races from mainland Cuba need to be
examined carefully and extensively. The Isla de
la Juventud population (L. c. microcyon), and to
some extent those from Little Cayman and Cay-
man Brac (L. c. granti), apparently differ in some
skeletal characters (rugose skull, incomplete over-
lap of nasal process of premaxilla). My samples
of these populations were inadequate to assess
accurately the significance of these differences.

Leiocephalus cubensis (Gray, 1840)

Tropidurus (Liolaemus) cubensis Gray, 1840.
Holotropis vittatus Hallowell, 1856.
Leiocephalus cubensis—Stejneger, 1917.
Holotype.-—BMNH XXIII.98a Type locality:
restricted by Schwartz (1959b) to the vicinity of
Guanabacoa, Habana Province, Cuba.
Distribution.—Cuba, island-wide but scat-
tered; Isla de la Juventud.
Characterization.—Large size (males to 120
mm, females to 85 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary
suture curved; nasal processes of frontal not ex-
posed; frontal narrow; parietal table U-shaped in
adults; supratemporal lateral; supratemporal pro-
cess of squamosal indistinct; skull roof rugose;
angular process of dentary well developed; tran-
sition to tricuspid crowns at Tooth 10 or 11 on
dentary; snout scale pattern Type II (3 rows be-
tween internasals and anterior frontals, with pos-
terior row composed of 3 scales in contact with
anterior frontals); parietal scale pattern Type II
(most lateral parietal scales subequal to or slightly
larger than median pair, and single row (occa-
sionally 2) of postparietals); 3 internasals, usu-
ally 2 in contact with rostral; 4 lorilabial scales
anterior to enlarged subocular; cephalic scale
ridges extend onto frontonasals; lateral trunk
scales not reduced; dorsal crest moderate, 48—64
dorsal crest scales occiput—vent; usually 4 postanal
escutcheons; tricarinate scales of first and second
toe enlarged, fringe moderate; lateral neck folds

moderate; nuchal fold moderately convex; lateral
fold on trunk absent; distinctive scapular and
suprascapular pattern absent; facial mask present.

Subspecies.—cubensis Gray; gigas Schwartz
(1959b); minor Garrido (in Varona and Garrido,
1970); pambasileus Schwartz (1959b); paraphrus
Schwartz (1959b);

Remarks.—Aggressive behavior in
Leiocephalus cubensis has been described by
Milera (1984), who observed a large male L.
cubensis repelling a rat that had fled into the
lizard’s burrow. The same male also was ob-
served seizing an adult male Anolis sagrei, pre-
sumably as a prey item.

Leiocephalus cuneus Etheridge, 1964

Holotype.—FSM 8226. Left dentary. Type
locality: Cave V, Two Foot Bay on the north coast
of Barbuda, BWI, by Clayton E. Ray and Robert
Allen, 31 March 1963.

Distribution.—Extinct and known only by
fossils from Barbuda and Antigua; possibly
Anguilla and Guadeloupe (see Remarks).

Characterization.—Large size (to 200 mm
SVL); transition from simple to tricuspid tooth
crowns complete at Tooth 4 or 5 on dentary,
Tooth 3 on maxilla; intramandibular septum
present.

Remarks.—Etheridge (1964) referred numer-
ous cranial and postcranial elements to this spe-
cies, which is known elsewhere on Barbuda from
the Indian Town Trail archaeological site (Watters
etal., 1984). Abundant remains were also discov-
ered in Burma Quarry fissure (late Holocene) on
the adjacent island of Antigua (Steadman et al.,
1984; Pregill et al., 1988). Fossils from Anguilla
and Guadeloupe also may belong to this species
(see above, and Pregill et al., 1988).

Leiocephalus endomychus new combination

Leiocephalus vinculum endomychus Schwartz,
1967a.

Holotype.—MCZ 81099. Type locality: 5.5
km NE Barrage de Péligre, 361 m, Dépt. del’Ouest
(de Centre), Haiti, 11 July 1962, by David C.
Leber and Albert Schwartz.

42 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Distribution.—Known only from the vicinity
of the type locality.

Characterization.—Moderate size (males to
69 mm, females to 56 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary
suture straight; nasal processes of frontal ex-
posed; frontal narrow; parietal table U-shaped in
adults; supratemporal lateral; supratemporal pro-
cess of squamosal indistinct; skull roof essen-
tially smooth; angular process of dentary well
developed; anterior transition to tricuspid crowns
at Tooth 3 or 4 on dentary; snout scale pattern
Type III (2 rows of scales between internasals
and anterior pair of frontals, with posterior row
composed of 3, or a single pair of, enlarged
scales); parietal scale pattern Type II (most lat-
eral parietal scales subequal to or slightly larger
than median pair, and single row [occasionally 2]
of postparietals); 3 internasals, usually 2 in con-
tact with rostral; 4 lorilabial scales anterior to
enlarged subocular; cephalic scales essentially
smooth; enlarged temporal scale usually present;
lateral neck scales keeled, undifferentiated; mid-
dorsal crest moderate, 51—65 dorsal crest scale
occiput—vent; postanal escutcheons numerous, in
3 or 4 rows; tricarinate scales of first and second
toe enlarged, fringe moderate; lateral neck folds
moderate; nuchal fold moderately convex; lat-
eral fold on trunk absent; distinctive scapular and
suprascapular pattern absent.

Remarks.—In his description of Leiocephalus
rhutidira, Schwartz (1979a) suggested that his
new taxon might be properly regarded as a sub-
species of L. vinculum. His impression was
strengthened by the similarities of L. rhutidira to
the mainland population, L. v. endomychus. Chro-
matic differences, slightly smaller size, and the
fact that juveniles and subadults display dorso-
lateral stripes were sufficiently compelling dif-
ferences to treat L. rhutidira as specifically dis-
tinct. Schwartz (1979a) also was impressed by
what he believed were conspicuous lateral neck
folds in L. rhutidira; hence, the etymology rhutis
(wrinkled) deire (neck). Leiocephalus vinculum
is polymorphic with respect to the complexity of
lateral neck folds (Character 31). Nominate vin-
culum on Ile de la Gonave have the simple con-
dition (State 2—gular and short antehumeral folds

only), whereas L. endomychus, like L. rhutidira,
possesses the moderate state (State 1—gular,
antehumeral and oblique folds). They differ fur-
ther from nominate L. vinculum in sharing a
straight-sided nasal-maxillary suture (Character
4.1, as in L. lunatus and L. semilineatus); ex-
posed nasal processes of the frontal (5.1, several
other species also); a laterally placed
supratemporal (9.0); the more anterior transition
to tricuspid teeth (13.1 unique); the presence
(usually) of an enlarged temporal scale dorsolat-
eral to the ear (23.1 as in barahonensis and
semilineatus), and the possession of multiple
escutcheon scales (29.2 unique). Leiocephalus
rhutidira differs from L. endomychus by having
striped juveniles and greater development of rug-
osities on the skull roof. These two taxa may be
conspecific, but additional specimens from inter-
vening localities in Haiti will need to be studied.
For the present, there is ample justification for
recognizing Leiocephalus endomychus apart from
L. vinculum, but less justification for placing L.
rhutidira in the synonymy of L. endomychus.

Leiocephalus eremitus Cope, 1868

Liocephalus (sic) eremitus Cope, 1868.
Holotype.—USNM 12016. Type locality:
Navassa Island, by W. J. Rasin.
Distribution.—Navassa Island, now extinct;
known only by the holotype.
Characterization. Moderate size (63 mm
SVL); angular process of dentary not well devel-
oped; transition to tricuspid crowns at Tooth 10
or 11 on dentary; snout scale pattern Type I (3,
rarely 4, rows subequal scales between internasals
and anterior frontals, none enlarged, posterior
row of at least 4 scales in contact with anterior
frontals); parietal scale pattern Type III (most
lateral parietals larger than median pair,
postparietals few or absent); 3 internasal scales,
usually all in contact with rostral; 6 lorilabial
scales anterior to enlarged subocular; head scales
smooth; lateral neck scales small, differentiated;
lateral trunk scales not reduced; middorsal crest
moderate, 50 dorsal crest scales occiput—vent;
tricarinate scales of first and second toe not con-
spicuously enlarged; lateral neck folds complex;

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 43

nuchal fold moderately convex; lateral fold on
trunk absent; distinctive scapular or suprascapular
pattern absent.

Remarks.—Characterization is based on the
only known specimen, a female bearing three
mature ova. A purported second specimen was
collected by the R. H. Beck expedition in 1917
and later illustrated by Schmidt (1921b). This
specimen (AMNH 16919), lacking locality data,
was reidentified correctly as L. melanochlorus by
Thomas (1966) following his visit to Navassa in
1965. Thomas (1966) also provided color notes
on the preserved holotype. Leiocephalus eremitus
is one of several species endemic to this small,
limestone island in the Jamaican Channel, about
60 km W Cap des Irois, Haiti.

Leiocephalus etheridgei Pregill, 1981

Holotype.-—USNM(VP) 259190. Right
dentary. Type locality: Blackbone | Cave, 1.2 km
S Barrio de Barahona, Municipio de Morovis,
Puerto Rico. Late Pleistocene.

Distribution.—Extinct, known only by fos-
sils from the type locality in Puerto Rico.

Characterization.—Large size (to 115 mm
SVL); acute, convex ridge present on anterolabial
face of dentary below the mental foramina; ante-
rior opening of Meckel’s groove extending from
the level of Tooth 6 forward to symphysis of jaw.

Remarks.—Pregill (1981) referred other
dentaries, cranial bones, and vertebrae to this
taxon, one of two species of Leiocephalus from
Puerto Rico known only by fossils.

Leiocephalus greenwayi Barbour and
Shreve, 1935

Holotype.—MCZ 36711. Type locality: East
Plana Cay, Bahamas.

Distribution.—Known only from East Plana
Cay.

Characterization.—Moderate size (males to
75mm SVL); nasal overlap of premaxillary spine
incomplete; nasal-maxillary suture curved; nasal
processes of frontal exposed between prefrontals
and nasals; frontal narrow; parietal table nar-
rowly constricted posteriorly, V-shaped in adults;

supratemporal lateral; supratemporal process of
squamosal indistinct; skull roof rugose; angular
process of dentary not well developed; transition
to tricuspid crowns at Teeth 7—9 on dentary;
snout scale pattern Type II (3 rows between
internasals and anterior frontals, with posterior
row composed of 3 scales in contact with anterior
frontals); enlarged pair of frontonasals; parietal
scale pattern Type III; (most lateral parietals
larger than median pair, postparietals few or ab-
sent); 3 internasal scales, usually all in contact
with rostral; 4 lorilabial scales anterior to en-
larged subocular; head scales smooth; lateral neck
scales small, differentiated; lateral trunk scales
reduced; middorsal crest reduced, 68—70 dorsal
crest scales occiput—vent; postanal escutcheons
absent; tricarinate scales of first and second toe
enlarged, fringe moderate; lateral neck folds
moderate; nuchal fold moderately convex; dark
supra-axillary and supra-inguinal blotches
present.

Remarks.—Little has been written about
Leiocephalus greenwayi since its original de-
scription. Clough and Pulk (1971) listed the spe-
cies in their synopsis of the vertebrate fauna and
vegetation of East Plana Cay. Schwartz (1967b)
mentioned, but did not treat, this species in his
review of other southern Bahamian Leiocephalus
(inaguae and psammodromus). The Plana Cays
(= French Cays) are isolated approximately 30
km E Acklins Island.

Leiocephalus herminieri (Duméril and
Bibron, 1837)

Holotropis herminieri Duméril and Bibron, 1837.
Leiocephalus herminieri—Boulenger, 1885.

Syntypes.—MNHN 1826, 2389, 6829. Type
locality: presumably Martinique (see Remarks).
The only other specimen reported in the literature
is a skeleton (BMNH 52.12.3.10) received from
Paris (Boulenger, 1885); it probably is a syntype
also.

Distribution.—Presumably Martinique (see
Remarks), now extinct (see Barbour, 1914).

Characterization.—Large size (140 mm
SVL); nasal overlap of premaxillary spine in-
complete; nasal-maxillary suture curved; nasal

44 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

processes of frontal exposed; septomaxilla re-
duced; frontal narrow; parietal table U-shaped in
adults; supratemporal lateral; supratemporal pro-
cess of squamosal indistinct; skull roof rugose;
angular process of dentary not well developed;
transition to tricuspid crowns at Tooth 10 or 11 on
dentary; snout scale pattern Type I (3, rarely 4,
rows subequal scales between internasals and
anterior frontals, none enlarged, posterior row of
at least 4 scales in contact with anterior frontals);
parietal scale pattern Type I (scales small, most
lateral parietals smaller than median pair, 2-4
rows postparietals); 4 internasal scales; 4 lorilabial
scales anterior to enlarged subocular; cephalic
scales smooth; lateral neck scales small, differ-
entiated; lateral trunk scales not reduced; ventral
scales keeled; middorsal crest prominent, scales
attenuate, 51 dorsal crest scales occiput-vent;
postanal escutcheons absent; tricarinate scales of
first and second toe enlarged, fringe moderate;
lateral neck folds complex; nuchal fold strongly
convex; lateral fold on trunk absent; base of tail
compressed; distinctive scapular or suprascapular
pattern absent.

Remarks.—Several herpetologists have com-
mented on the confusion surrounding Martinique
as the type locality for Leiocephalus herminieri
and other West Indian reptiles (Boulenger, 1885;
Stejneger, 1904; Barbour, 1914, 1915; Etheridge,
1964; Baskins and Williams, 1966; Schwartz and
Thomas, 1975). The original data accompanying
MNHN 2389 indicated that it was collected on
“Trinite” by L’Herminier, who sent the specimen
to Paris. Whereas “Trinite” could be, and has
been (e.g., Barbour, 1914), interpreted as Trinidad,
there is a town of Trinité on the north coast of
Martinique. Thus, it is not clear whether
Leiocephalus herminieri was collected on
Trinidad or near Trinité, Martinique, or perhaps
from both places. The other two specimens,
MNHN 1826 and 6829, were collected by Guyon
and Plee and are also presumed to have come
from Martinique. Martinique was the most active
shipping port in the Lesser Antilles during the
nineteenth century and, for much of Plee’s mate-
rial, was merely the transfer point from the field
to museums in Europe. For example, Ronald
Crombie (in litt.) noted that “A brief search of

taxa described from Plee’s material supposedly
coming from Martinique yields the following:
Hylodes martinicensis, Ameiva major, A. pleei,
Anolis chlorocyanus, A. cristatellus, A.
marmoratus A. pulchellus, Celestus pleei,
Gonatodes albogularis, Sphaerodactylus fantas-
ticus, and Amphisbaena caeca.” In other words,
Plee’s collections included species from at least
Guadeloupe, Puerto Rico, Hispaniola, and the St.
Martin Bank. Some specimens did in fact come
from Martinique (Anolis roquet), but there is
nothing that assures us that Leiocephalus
herminieri was among these, or is endemic to that
island.

A fifth specimen of Leiocephalus herminieri
was discovered recently by Uno Svensson (pers.
comm., 1988) inthe Rijksmuseum van Natuurlijke
Historie, Leiden. Quite regrettably, it has since
been misplaced following spirit removals in the
collection (Marinus Hoogmoed, pers. comm.,
1989). According to Svensson, the label associ-
ated with the specimen read: “Liocephalus
herminieri, RMNH 2888, Coll: |’ Herminier, Mus.
Paris, 1835, Trinidad (?), old no: 166.” Svensson
measured the snout-vent length at 99 mm, and
counted 52 scales around midbody; the ventral
scales were keeled. Apparently, 1’ Herminier must
have collected two specimens, one of which was
sent to Paris and the other to Leiden.

This is the only species of Leiocephalus that
possesses keeled ventral scales.

Leiocephalus inaguae Cochran, 1931

Holotype.—USNM 81277. Type locality: Man
of War Bay, Great Inagua Island, 08 August 1930,
by P. Bartsch.

Distribution.—Great Inagua Island, Bahamas.

Characterization.—Moderately large size
(males to 90 mm, females to 74 mm SVL); nasal
overlap of premaxillary spine incomplete; pre-
maxillary spine broad at base, constricted, lateral
spike above constriction; nasal-maxillary suture
curved; nasal process of frontal exposed; frontal
wide, flat posteriorly; parietal table narrowly
constricted posteriorly, V-shaped in adults;
supratemporal lateral; supratemporal process of
squamosal indistinct; skull roof smooth; angular

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 45

process of dentary not well developed; transition
to tricuspid crowns at Teeth 7-9 on dentary;
snout scale pattern Type II (3 rows between
internasals and anterior frontals, with posterior
row composed of 3 scales in contact with anterior
frontals); parietal scale pattern Type I (most lat-
eral parietals larger than median pair, postparietals
few or absent); 3 internasal scales, usually 2 in
contact with rostral; 4 lorilabial scales anterior to
enlarged subocular; cephalic scales smooth; lat-
eral neck scales small, differentiated; lateral trunk
scales reduced; dorsal crest moderate, 65-77
dorsal crest scales occiput—vent; usually 4 postanal
escutcheons; tricarinate scales of first and second
toe enlarged, fringe moderate; lateral neck folds
complex; nuchal fold strongly convex; lateral
fold on trunk; suprascapular blotches conspicu-
ous, continuing down sides and fading.

Remarks.—Aside from the original and then
subsequent descriptions by Cochran (1931;
1934c), and later by Schwartz (1967b), there are
behavioral and natural history notes on this spe-
cies by Noble and Klingel (1932), which were
based on Klingel’s observations when he was
shipwrecked on Inagua during the winter of 1930—
31 (Klingel, 1932; 1941).

Leiocephalus jamaicensis Etheridge, 1966b

Holotype.-—AMNH(VP) 2311. Left dentary.
Type locality: Dairy Cave, 2.5km W Dry Harbour,
St. Ann Parish, Jamaica.

Distribution.—Extinct, known only by fos-
sils from Jamaica. Additional fossils are known
from Montego Bay Airport Cave at the west end
of the air strip, Montego Bay, St. James Parish,
and from Portland Ridge Caves, Clarendon Par-
ish (Etheridge, 1966b). Most recently reported
by Pregill et al. (1991) from Marta Tick Cave, 8
km WNW Quickstep, Trelawny Parish.

Characterization.—Large size (to 130 mm
SVL); pterygoid teeth absent; skull moderately
rugose; Meckel’s groove closed, fused; anterior
border of angular process of articular bone form-
ing an obtuse angle with medial border of articu-
lar; main axis of articular process projecting
medially from retroarticular condyle (Etheridge,
1966b).

Remarks.—Of the numerous cranial elements
and vertebrae referred to this species by Etheridge
(1966b), two frontal bones differ from one an-
other in two important ways. The Portland Cave
specimen (UF 8496) is wide posteriorly and bears
well-developed rugosities. In contrast, the fron-
tal from Montego Bay Airport Cave (UF 8508) 1s
plesiomorphic, being narrow and smooth. Al-
though ontogeny can influence these characters,
the two frontals are comparable in size. I think
that they represent two species, but which of
them is L. jamaicensis cannot be determined; the
holotype left dentary (AMNH 2311) came from
yet a different locality, Dairy Cave. Evidently, L.

jJamaicensis, whether one or two species, was

widespread over the central and western part of
the island and may have persisted into historical
times. The dentaries reported from Marta Tick
Cave by Pregill et al. (1991) were unmineralized
bones collected from the surface of the cave
floor; other fossils that were recovered with them
are associated with a C14 age of 770 + 70 ybp.
None of the Type material is believed to be older
than latest Pleistocene (Etheridge, 1966b).

Leiocephalus loxogrammus Cope, 1887

Syntypes.—USNM 14569 (3 specimens);
MCZ 10931. Type locality: Rum Cay, Bahama
Islands, by C. H. Townsend, J. E. Benedict, and
Fisher.

Distribution.—Rum Cay and San Salvador
(Watling’s) Island, Bahamas; possibly Concep-
tion Island, Bahamas (see Remarks).

Characterization.— Moderate size (males to
92 mm, females to 74 mm SVL); nasal overlap of
premaxillary spine incomplete; nasal maxillary
suture curved; nasal processes of frontal exposed;
septomaxilla reduced; frontal narrow; parietal
table narrowly constricted posteriorly, V-shaped
in adults; supratemporal lateral or ventomedial;
supratemporal process of squamosal distinct in
some; skull roof smooth; angular process of
dentary well developed; transition to tricuspid
crowns at Teeth 7—9 on dentary; neural processes
of trunk vertebrae obtuse, not expanded distally
in some; hypapophyses of trunk vertebrae nar-
row; snout scale pattern Type III (2 rows of scales

46 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

between internasals and anterior pair of frontals,
with posterior row composed of 3, or a single pair
of, enlarged scales); parietal scale pattern Type II
(most lateral parietal scales subequal to or slightly
larger than median pair, and single row [occasion-
ally 2] of postparietals); enlarged lateral postparietal
scale present; 3 internasal scales, usually all in
contact with rostral; 4 lorilabial scales anterior to
enlarged subocular; cephalic scales ridges restricted;
2 or 3 elongate temporal scales behind eye; lateral
neck scales keeled, undifferentiated; lateral trunk
scales not reduced; middorsal crest moderate, 55—
65 dorsal crest scales occiput—vent; usually 4
postanal escutcheons; tricarinate scales of first and
second toe enlarged, fringe moderately developed;
lateral neck folds moderate; nuchal fold moder-
ately complex; lateral fold on trunk absent; facial
band present.

Subspecies.—/oxogrammus Cope (Rum Cay);
parnelli Barbour and Shreve (1935) (San Salvador).

Remarks.—The type series was collected by
a party of the U.S. Fish Commission during the
cruise of the steamer Albatross, which stopped at
San Salvador as well as Rum Cay. There is no
compelling reason to doubt Rum Cay as the type
locality (Cope, 1887:438), but it is odd that the
expedition did not collect Leiocephalus loxo-
grammus on San Salvador where they obtained
Sphaerodactylus and Leptotyphlops. One expla-
nation is that the party confined its field activity
to the leeward side in the vicinity of Cockburn
Town, the only port on San Salvador. If so, their
failure to collect L. loxogrammus accords with
distributional records of Olson et al. (1990), who
found that the species was nearly restricted to the
opposite (windward) side of the island. They also
reported Holocene fossils of L. loxogrammus
from several sinkholes located in the southern
and eastern sections of San Salvador.

A population of Letocephalus loxogrammus
may exist on Conception Island northwest of
Rum Cay, but as yet, it has not been verified with
specimens (Schwartz et al., 1978).

Leiocephalus lunatus Cochran, 1934a

Leiocephalus personatus lunatus Cochran, 1934a.
Leiocephalus lunatus—Schwartz, 1967a.

Holotype.—FMNH 166. Type locality: Santo
Domingo, Distrito Nacional, Republica
Dominicana.

Distribution.—Southern and eastern Domini-
can Republic, primarily coastal; Isla Saona, Isla
Catalina.

Characterization.—Small to moderate size
(males to 67 mm, females to 60 mm SVL); nasal
overlap of premaxillary spine complete or not;
nasal-maxillary suture straight; nasal processes
of frontal not exposed; frontal narrow; parietal
table U-shaped in adults; supratemporal lateral;
supratemporal process of squamosal indistinct;
skull roof rugose; angular process of dentary not
well developed; transition to tricuspid crowns at
Teeth 7—9 on dentary; hypapophyses of posterior
trunk vertebrae usually narrow; snout scale pat-
tern Type IH (2 rows of scales between internasals
and anterior pair of frontals, with posterior row
composed of 3, or a single pair of, enlarged
scales); parietal scale pattern Type II (most lat-
eral parietal scales subequal to or slightly larger
than median pair, and single row [occasionally 2]
of postparietals); enlarged lateral postparietal
scale present; 3 internasal scales, usually 2 in
contact with rostral; 4 lorilabial scales anterior to
enlarged subocular; cephalic scale ridges re-
stricted; moderately enlarged temporal scale pres-
ent in some; lateral neck scales keeled,
undifferentiated; lateral trunk scales not reduced;
middorsal crest prominent, scales attenuate, 50—
68 dorsal crest scale occiput—vent; usually 4
postanal escutcheons; tricarinate scales of first
and second toe enlarged, fringe moderate; lateral
neck folds moderate; nuchal fold moderately
convex; lateral fold on trunk absent; supraaxillary
blotch in males; distinct spotting on chin and
throat.

Subspecies.—/unatus, Cochran (1934a);
arenicolor Mertens (1939); lewisi Schwartz
(1967a); louisae Cochran (1934b); melaenascelis
Schwartz (1967a); thomasi Schwartz (1967a).

Remarks.—Considerable variation in skel-
etal and integumentary (squamation) characters
calls for a closer examination of this species.
The most detailed descriptions are those of
Cochran (1934a,b; 1941) and especially
Schwartz (1967a).

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 47

Leiocephalus macropus Cope, 1863

Liocephalus (sic) macropus Cope, 1863.
Leiocephalus macropus—Stejneger, 1917.

Lectotype—USNM 25819, selected by Hardy
(1958a); syntypes MCZ 10930; USNM 12254,
25819-23, 25825-29 (see Remarks). Type local-
ity: restricted by Stejneger (1917) to Monte Verde,
Guantanamo Province, Cuba (but see Remarks).

Distribution.—Cuba, from Pinar del Rio east,
but primarily eastern Cuba.

Characterization Moderate to large size
(males to 92 mm, females to 75 mm SVL); nasal
overlap of premaxillary spine complete; base of
nasal process of premaxilla broad; nasal-maxil-
lary suture curved; nasal processes of frontal
exposed; frontal broad and flat posteriorly; pari-
etal table U-shaped in adults; supratemporal lat-
eral; supratemporal process of squamosal usu-
ally distinct; skull roof smooth; angular process
of dentary not well developed; transition to tri-
cuspid crowns at Tooth 10 or 11 on dentary; snout
scale pattern Type II (3 rows between internasals
and anterior frontals, with posterior row com-
posed of 3 scales in contact with anterior frontals);
parietal scale pattern Type I (scales small, most
lateral parietals smaller than median pair, 2-4
rows postparietals); usually 2 internasal scales; 4
lorilabial scales anterior to enlarged subocular;
cephalic scale ridges extend onto frontonasals;
lateral neck scales small, differentiated; lateral
trunk scales reduced; middorsal crest moderate,
50-73 dorsal crest scales occiput—vent; usually 4
postanal escutcheons; tricarinate scales of first
and second toe enlarged, fringe moderate; lateral
neck folds complex; nuchal fold strongly con-
vex; lateral fold on trunk present; supraaxillary
blotch present, bisected by white vertical bar;
dark facial band extending onto trunk.

Subspecies.—macropus Cope (1863); aegialus
Schwartz and Garrido (1967); asbolomus Schwartz
and Garrido (1967); felinoi Garrido (1979); hoplites
Zug (1959); hyacinthurus Zug (1959); immaculatus
Hardy (1958a); koopmani Zug (1959); lenticulatus
Garrido (1973b); phylax Schwartz and Garrido
(1967); torrei Garrido (1979).

Remarks.—Schwartz and Garrido (1967)
determined that the presumed syntypes (Cochran,

1961) of Leiocephalus macropus might not have
been the material Cope had in front of him when
he described this species. For example, the sup-
posed syntypes also included a specimen of L.
raviceps, which Cope (1863) himself assuredly
could have distinguished because he described
that species in the same paper. Moreover, Hardy’s
(1958a) selection of USNM 25819 as the lecto-
type was based on the assertion that that speci-
men best fit Cope’s description. However, this is
not the case, which casts further doubt on the
veracity of the type series. Hence, restricting the
type locality to Monte Verde (Stejneger, 1917)
becomes moot, as the USNM series likely was
collected from several stations in eastern Cuba
(Schwartz and Garrido, 1967).

Leiocephalus macropus displays considerable
clinal variation in pattern and scutellation through-
out its range (see also Hardy, 1958b).

Leiocephalus melanochlorus Cope, 1863

Syntypes.—MCZ 3598; USNM 53402; CAS
39392. Type locality: near Jérémie, Dépt. de la
Grand’ Anse, Haiti.

Distribution.—Southern Haiti, essentially the
Tiburon Peninsula; Ie-a- Vache.

Characterization.—Large size (males to 130
mm, females to 102 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary
suture straight; nasal processes of frontal ex-
posed; septomaxilla reduced; frontal broad and
flat posteriorly; parietal table U-shaped in adults;
supratemporal lateral; supratemporal process of
squamosal distinct; skull roof smooth; angular
process of dentary not well developed; transition
to tricuspid crowns at Tooth 10 or 11 on dentary;
snout scale pattern Type I (3, rarely 4, rows
subequal scales between internasals and anterior
frontals, none enlarged, posterior row of at least
4 scales in contact with anterior frontals); pari-
etal scale pattern Type I (scales small, most lat-
eral parietals smaller than median pair, 2—4 rows
postparietals); usually 2 internasal scales; 6
lorilabial scales anterior to enlarged subocular;
cephalic scale ridges extend onto frontonasals;
body scales large; lateral neck scales small, dif-
ferentiated; lateral trunk scales reduced; middor-

48 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

sal crest moderate, 37-53 dorsal crest scales
occiput—vent; postanal escutcheons absent;
tricarinate scales of first and second toe enlarged,
fringe moderate; lateral neck folds complex; nu-
chal fold strongly convex; lateral fold on trunk
present; suprascapular blotches present, continu-
ing onto trunk; tail compressed at base.

Subspecies.—melanochlorus Schwartz
(1965); hypsistus Schwartz (1965).

Leiocephalus onaneyi Garrido, 1973a

Holotype.—IZ-2869. Type locality: the top
of Loma de Mocamba, between San Antonio del
Sur and Imias, Oriente (Guantanamo) Province,
Cuba.

Distribution.—Known only from the type
locality, but presumed to occur in the hills of the
Sierra de Imias.

Characterization Moderate size (73 mm
SVL); belly and throat white, without spots, re-
ticulations, or marks; dorsum with eight well-
delimited zones of alternating white and choco-
late; (median) parietal scales only in contact
posteriorly; supraorbital semicircles incomplete;
48 dorsal crest scales occiput—vent; five loreals;
4 or 5 supralabials and an equal number of
infralabials.

Remarks.—Besides the holotype, an adult
female, only two other specimens of Leiocephalus
onaneyi have been reported, the paratypes listed
by Garrido—a male (IZ 2848) and juvenile fe-
male (IZ 2870). The characterization above is
taken directly from Garrido’s diagnosis and, there-
fore, is not comparable to other species accounts
in this section. In his comparison with other
Cuban Leiocephalus, Garrido (1973a) commented
that only L. stictigaster would likely be confused
with L. onaneyi, but that they differed in the
immaculate throat and the distinct dorsum of
eight zones (6 less distinct zones in L. stictigaster).
From the accompanying photographs in Garrido
(1973a), L. onaneyi indeed gives the impression
of an aberrant L. stictigaster. The cephalic scales
seem to have well-defined ridges extending onto
the frontonasals.

Leiocephalus onaneyi is restricted to a small
region of sharp karst topography characterized

by low rainfall and a high level of endemic,
xerophytic vegetation.

Leiocephalus partitus Pregill, 1981

Holotype.—USNM(VP) 259203. Right
dentary. Type locality: Guanica Bat Cave, Reserva
Forestal Guanica, 6 km E Barrio de Guanica,
Municipio de Guayanilla, Puerto Rico.

Distribution.—Extinct, known only by fos-
sils from the type locality, and from Cueva del
Perro, Municipio de Morovis, Puerto Rico
(Pregill, 1981).

Characterization.—Large size (estimated
SVL 125-130 mm); dentary with a well-devel-
oped intramandibular septum; tooth crowns nar-
row and weakly flared; Meckel’s groove closed
and fused except from below the seventh tooth to
the anterior tip of the jaw.

Remarks.—Only two fossils of this species
are known, the holotype and another right dentary
(KUVP 11473) from Cueva del Perro. The two
fossil localities are on opposite sides of the island
from one another in northcentral and southwest
Puerto Rico, respectively. Although no radiocar-
bon ages are available for either of these deposits,
they are most likely late Pleistocene to Holocene
age (Pregill, 1981).

Leiocephalus personatus Cope, 1863

Liocephalus (sic) trigeminatus Cope, 1863.
Leiocephalus personatus—Cochran, 1932 (part).
Leiocephalus personatus—Schwartz, 1967a.

Syntype.—MCZ 3615. Type locality: near
Jérémie, Dépt de la Grand’ Anse, Haiti. The holo-
type of Leiocephalus trigmenatus is lost (for-
merly in MCZ).

Distribution.—Island-wide on Hispaniola,
especially northern Dominican Republic.

Characterization.—Moderate size (males to
86 mm, females to 63 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary su-
ture curved; nasal process of frontal exposed in
some; frontal narrow; parietal table U-shaped in
adults; supratemporal lateral; supratemporal pro-
cess of squamosal indistinct; skull roof rugose;
angular process of dentary well developed; tran-

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 49

sition to tricuspid crowns at Tooth 10 or 11 on
dentary; hypapophyses of trunk vertebrae usu-
ally narrow; snout scale pattern Type III (2 rows
of scales between internasals and anterior pair of
frontals, with posterior row composed of 3, or a
single pair of, enlarged scales); parietal scale
pattern Type II (most lateral parietal scales
subequal to or slightly larger than median pair,
and single row [occasionally 2] of postparietals);
3 internasal scales, usually 2 in contact with
rostral; 4 lorilabial scales anterior to enlarged
subocular; cephalic scale ridges extend onto
frontonasals; moderately enlarged temporal scale
present in some; lateral neck scales undiffer-
entiated: lateral trunk scales not reduced; mid-
dorsal crest prominent, scales attenuate, 41-64
dorsal crest scales occiput—vent; usually 4 postanal
escutcheons; tricarinte scales of first and second
toe enlarged, fringe moderately developed:
lateral neck folds moderate; nuchal fold moder-
ately convex; antegular scale fold present in some;
lateral fold on trunk absent; ventral pattern ab-
sent in males, females often with dark spotting on
throat and chest; facial mask usually evident.

Subspecies.—personatus Cope (1863), Barbour
(1935, with which Barbour was tempted to lump L.
herminieri); actites Schwartz (1967a); agraulus
Schwartz (1967a); budeni Schwartz (1967a)
elattoprosopon Gali, Schwartz, and Suarez (1988);
mentalis Cochran (1932); poikilometes Schwartz
(1969): pyrrholaemus Schwartz (1971); scalaris
Cochran (1932); pulcherrimus Mertens (1939);
socoensis Gali and Schwartz (1982); tarachodes
Schwartz (1967a): trujilloensis Mertens (1939).

Remarks.—Leiocephalus personatus is an
especially variable species, as demonstrated by
the numerous populations given subspecific des-
ignation, and in the character discordance noted
above. A more detailed examination of this wide-
spread Hispaniolan taxon is warranted.

Leiocephalus pratensis (Cochran, 1928a)

Hispaniolus pratensis Cochran, 1928a.
Leiocephalus pratensis Etheridge, 1966a.
Holotype.—USNM 69189. Type locality:
Atalaye Plantation near St. Michel, Dépt du Nord,
Haiti; emended by Schwartz (1968) to Atalaye

Plantation near St.-Michel de |’ Atalaye, Dépt de
1’ Artibonite, Haiti.

Distribution.—Known only from the vicinity
of the type locality, and from Ile a Cabrit in the
Golfe de la Gonave.

Characterization.—Small to moderate size
(males to 64 mm, females to 55 mm SVL); nasal
overlap of premaxillary spine complete: nasal-
maxillary suture curved; nasal processes of fron-
tal not exposed; frontal wide and flat posteriorly;
parietal table U-shaped in adults; supratemporal
lateral; supratemporal process of squamosal in-
distinct; skull roof smooth; angular process of
dentary not well developed; transition to tricus-
pid crowns at Tooth 10 or 11 on dentary; snout
scale pattern Type III (2 rows of scales between
internasals and anterior pair of frontals, with
posterior row composed of 3, or a single pair of,
enlarged scales); parietal scale pattern Type HI
(most lateral parietals larger than median pair,
postparietals few or absent); usually 2 internasal
scales; 4 lorilabial scales anterior to enlarged
subocular; supraocular scales usually 5/5; ce-
phalic scale ridges restricted; lateral neck scales
keeled, undifferentiated; lateral trunk scales not
reduced: middorsal crest absent; usually 4 postanal
escutcheons; lenticular scale organs absent;
tricarinate scales of first and second toe not en-
larged, fringe weakly developed; lateral neck
folds simple; nuchal fold transverse; lateral fold
on trunk absent; distinctive scapular or
suprascapular pattern absent.

Subspecies.—pratensis, Cochran (1928a);
chimarus Schwartz (1979b).

Leiocephalus psammodromus Barbour, 1916a

Liocephalus (sic) arenarius Barbour, 1916a.
Leiocephalus psammodromus Barbour, 1920
(substitute name for arenarius, preoccupied
by Steironotus [Ophryoessoides| arenarius
Tschudi, 1845).
Holotype.—MCZ 11948. Type locality: “Bas-
tion Cay,” Turks Island, B.W.I., by L. L. Mowbry.
Distribution.—Turks and Caicos Islands; scat-
tered on associated Cays.
Characterization.—Large size (males to 105
mm, females to 84 mm SVL); nasal overlap of

50 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

premaxillary spine incomplete; premaxillary spine
broad, constricted at base and with spikelike
lateral process; nasal-maxillary suture curved;
nasal processes of frontal exposed; septomaxilla
reduced; frontal broad and flat posteriorly; pari-
etal table narrowly constricted posteriorly, V-
shaped in adults; supratemporal ventromedial;
supratemporal process of squamosal distinct; skull
roof smooth; angular process of dentary not well
developed; transition to tricuspid crowns at Tooth
10 or 11 ondentary; snout scale pattern Type I (3,
rarely 4, rows subequal scales between internasals
and anterior frontals, none enlarged, posterior
row of at least 4 scales in contact with anterior
frontals); parietal scale pattern Type I (scales
small, most lateral parietals smaller than me-
dian pair, 2-4 rows postparietals); usually 4
internasal scales; 5 or 6 lorilablial scales ante-
rior to enlarged subocular; cephalic scales es-
sentially smooth; lateral neck scales small, dif-
ferentiated; lateral trunk scales reduced middor-
sal crest moderate, 56—81 dorsal crest scales
occiput—vent; body scales small; 2—4 postanal
escutcheons occasionally present; tricarinate
scales of first and second toe enlarged, fringe
moderate; lateral neck folds complex; nuchal
fold strongly convex; lateral fold on trunk pres-
ent; suprascapular blotches present; base of tail
laterally compressed.

Subspecies.—psammodromus Barbour
(1920), Schwartz and Thomas (1975); aphretor
Schwartz (1967b); apocrinus Schwartz (1967b);
cacodoxus Schwartz (1967b); hyphantus
Schwartz (1967b); mounax Schwartz (1967b).

Remarks.—The type locality, “Bastion Cay,”
cannot be located on maps. Leiocephalus
psammodromus is a variable species that would
benefit from further taxonomic study. Holocrine
glands of this species were described by Alexander
and Maderson (1972).

Leiocephalus punctatus Cochran, 1931

Leiocephalus carinatus punctatus Cochran, 1931.

Leiocephalus carinatus helenae Barbour and
Shreves1935:

Leiocephalus carinatus picinus Barbour and
Shreve, 1935.

Leiocephalus punctatus—Etheridge, 1966a (as
suggested by Rabb, 1957).
Holotype.-—USNM 81560. Type locality:

North shore of the bay at Jamaica Wells, Acklin’s

Island, Bahamas.
Distribution.—Crooked-Acklins

Samana (Atwood) Cay.
Characterization.— Moderate size (males to

78 mm SVL); nasal overlap of premaxillary spine

incomplete; premaxillary spine broad, constricted

at base; nasal-maxillary suture curved; nasal pro-
cesses of frontal exposed; frontal broad and flat
posteriorly; parietal table narrowly constricted
posteriorly, V-shaped in adults; supratemporal
ventromedial; supratemporal process of squamo-
sal indistinct; skull roof smooth; angular process
of dentary not well developed; transition to tri-
cuspid crowns at Tooth 10 or 11 on dentary; snout
scale pattern Type II (3 rows between internasals
and anterior frontals, with posterior row com-
posed of 3 scales in contact with anterior frontals);
enlarged median pair of frontonasals; parietal
scale pattern Type II (most lateral parietals larger

than median pair, postparietals few or absent); 3

internasals, usually 2 in contact with rostral; 5 or

6 lorilabial scales anterior to enlarged subocular:

cephalic scales smooth; moderately enlarged tem-

poral scale in some; lateral neck scales small,
differentiated; lateral trunk scales not reduced;
middorsal crest moderate, 49-59 dorsal crest
scales occiput—vent; postanal escutcheons ab-
sent; tricarinate scales of first and second toe
enlarged, fringe moderate; lateral neck folds mod-
erate; nuchal fold moderately convex; lateral
fold on trunk absent; scapular or suprascapular
pattern absent; distinctive light spotting on head.

Remarks.—Meristic and morphometric de-
tails of this species were treated by Rabb (1957).

Bank,

Leiocephalus raviceps Cope, 1863

Syntypes.—ANSP 8601-03; MCZ 10928;
USNM 4162. Type locality: Eastern Cuba; re-
stricted by Gundlach (1880) to the mountains
near Guantanamo, Oriente.

Distribution.—Primarily eastern half of Cuba,
but disjunct populations in Matanzas and Pinar
del Rio Provinces.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 51

Characterization.—Moderate size (males to
72mm, females to60 mm SVL); nasal overlap of
premaxillary spine complete; nasal maxillary
suture curved; nasal processes of frontal not ex-
posed; frontal narrow; parietal table U-shaped in
adults; supratemporal lateral; supratemporal pro-
cess of squamosal indistinct; skull roof rugose:
angular process of dentary well developed; tran-
sition to tricuspid crowns at Tooth 10 or 11 on
dentary; neural processes of trunk vertebrae at
obtuse angle, not distally expanded; hypapophyses
of trunk vertebrae narrow; snout scale pattern
Type II (2 rows of scales between internasals
and anterior pair of frontals, with posterior row
composed of 3, or a single pair of, enlarged
scales); parietal scale pattern Type II (most lat-
eral parietal scales subequal to or slightly larger
than median pair, and single row [occasionally 2]
of postparietals); enlarged postparietal scale lat-
erally; 3 internasals, usually 2 in contact with
rostral; 4 lorilabial scales anterior to enlarged
subocular; cephalic scale ridges extend onto
frontonasals; 2 or 3 elongate temporal scales
behind eye; lateral neck scales keeled undiffer-
entiated; lateral trunk scales not reduced; mid-
dorsal crest moderate, 55—74 dorsal crest scales
occiput—vent; usually 4 postanal escutcheons;
tricarinate scales of first and second toe enlarged,
fringe moderate; lateral neck folds moderate;
nuchal fold moderately convex; lateral fold
present on trunk; dark facial band present, ex-
tending onto trunk.

Subspecies.—raviceps Cope (1863), Schwartz
(1960b); delavarai Garrido (1973b); jaumei
Schwartz and Garrido (1968b):; klinikowski
Schwartz (1960b): uzzelli Schwartz (1960b).

Leiocephalus rhutidira Schwartz, 1979a

Holotype.—CM 60520. Type locality:
Lapierre, 10.6 km W Ca Soleil, Dépt. de
l’Artibonite, Haiti.

Distribution.—Known only from the type
locality.

Characterization.—Small size (males to 66
mm, females to 57 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary
Suture straight; nasal processes of frontals ex-

posed; frontal narrow; parietal table U-shaped in
adults; supratemporal lateral; supratemporal pro-
cess of squamosal indistinct; skull roof rugose;
angular process of dentary well developed; tran-
sition to tricuspid crowns at Tooth 3 or 4 on
dentary; snout scale pattern Type III (2 rows of
scales between internasals and anterior pair of
frontals, with posterior row composed of 3, or a
single pair of, enlarged scales); parietal scale
pattern Type II (most lateral parietal scales
subequal to or slightly larger than median pair,
and single row [occasionally 2] of postparietals):
3 internasals, usually 2 in contact with rostral; 4
lorilabial scales anterior to enlarged subocular;
cephalic scales essentially smooth; enlarged tem-
poral scale usually present; lateral neck scales
keeled, undifferentiated; lateral trunk scales not
reduced; middorsal crest moderate, 51—65 dorsal
crest scale occiput—vent; postanal escutcheons
numerous in 3 or 4 rows; tricarinate scales of first
and second toe enlarged, fringe moderate; lateral
neck folds moderate; nuchal fold moderately
convex; lateral fold on trunk absent; scapular or
suprascapular pattern absent; juveniles with
striped dorsum.

Remarks.—See account of Leiocephalus
endomychus.

Leiocephalus schreibersi (Gravenhorst, 1837)

Pristinotus schreibersii Gravenhorst, 1837.
Steironotus schreibersi—Fitzinger, 1843.
Leiocephalus schreibersi—Cope, 1868.
Holotype.—Unlocatable, apparently depos-
ited in Breslau Museum; Type locality: “San
Domingo”; restricted by Schwartz (1968) to the
vicinity of Port-au-Prince, Dépt. de 1’ Ouest, Haiti.
Distribution.—Fairly widely scattered north
to south in central Hispaniola, but absent from
large areas of the interior, west to Port-au-Prince;
Ile de la Tortue; introduced in south Florida.
Characterization.—Moderately large size
(males to 107 mm females to 78 mm SVL); nasal
overlap of premaxillary spine complete, premax-
illary spine wide, constricted at base; nasal-max-
illary suture curved; nasal processes of frontal
not exposed; frontal narrow; parietal table U-
shaped in adults; supratemporal lateral; supra-

a2 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

temporal process of squamosal indistinct; skull
roof smooth; angular process of dentary well
developed; transition to tricuspid crowns at Tooth
10 or 11 on dentary; neural processes of trunk
vertebrae at obtuse angle, not expanded distally;
clavicle narrow; snout scale pattern Type II (3
rows between internasals and anterior frontals,
with posterior row composed of 3 scales in con-
tact with anterior frontals); parietal scale pattern
Type I (scales small, most lateral parietals smaller
than median pair, 2-4 rows postparietals); 3
internasal scales, usually 2 in contact with ros-
tral; 4 lorilabial scales anterior to enlarged
subocular; cephalic scale ridges restricted; lat-
eral neck scales small, differentiated; lateral trunk
scales reduced; middorsal crest small, 63-87
dorsal crest scales occiput—vent; usually 4 postanal
escutcheons; tricarinate scales of first and second
toe enlarged, fringe moderate; lateral neck folds
complex; nuchal fold strongly convex; lateral
fold present on trunk; 5—7 transverse rows of
dark, contrasting scales on venter; scapular or
suprascapular pattern absent.

Subspecies.—schreibersi Gravenhorst (1837),
Schwartz (1968); nesomorus Schwartz (1968).

Remarks.—Aspects of the thermal biology of
L. schreibersi were described by Marcellini and
Jenssen (1989).

Leiocephalus semilineatus Dunn, 1920

Leiocephalus personatus semilineatus—Cochran,
1941.
Leiocephalus semilineatus—Schwartz,1967a,
1968.
Holotype.—MCZ 12748. Type locality:
Thomazeau, Dépt de l'Ouest, Haiti.
Distribution.—Southcentral Hispaniola, west
to Port-au-Prince, east to near Bani.
Characterization.—Small size (males to 53
mm, females to 48 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary
suture straight; nasal processes of frontal not
exposed; frontal narrow; parietal table U-shaped
in adults; supratemporal lateral; supratemporal
process of squamosal indistinct; skull roof rugose;
angular process of dentary well developed; tran-
sition to tricuspid crowns at Tooth 10 or 11 on

dentary; snout scale pattern Type III (2 rows of
scales between internasals and anterior pair of
frontals, with posterior row composed of 3, or a
single pair of, enlarged scales); parietal scale
pattern Type II (most lateral parietal scales
subequal to or slightly larger than median pair,
and single row [occasionally 2] of postparietals);
3 internasals, usually 2 in contact with rostral; 4
lorilabial scales anterior to enlarged subocular;
cephalic scales essentially smooth; single, en-
larged temporal scale usually present; lateral neck
scales undifferentiated; lateral trunk scales not
reduced; middorsal crest moderate, 49-63 dorsal
crest scales occiput—vent; postanal escutcheons
usually 4; tricarinate scales of first and second toe
enlarged, fringe moderate; lateral neck folds
moderate; nuchal fold moderately convex; lat-
eral fold on trunk absent; facial band extending
onto trunk; throat and venter immaculate.

Remarks.—Williams (1963) noted the asso-
ciation of this species with other xerophilous
lizards, for example Anolis whitemani, in the
open scrub of the Cul de Sac Plain and Valle de
Neiba.

Leiocephalus stictigaster Schwartz, 1959b

Holotropis microlophus Duméril and Bibron,
1837 (part).
Leiocephalus vittatus Boulenger, 1885 (part).
Leiocephalus cubensis Barbour, 1916b (part).
Holotype.-—AMNH 77864. Type locality:
Beach on Cabo Corrientes, Pinar del Rio Prov-
ince, Cuba
Distribution.—Cuba island-wide, but scat-
tered; Isla de la Juventud.
Characterization.—Moderately large size
(males to 100 mm, females to 80 mm SVL); nasal
overlap of premaxillary spine complete; nasal-
maxillary suture curved; nasal processes of fron-
tal not exposed; frontal narrow; parietal table U-
shaped in adults; supratemporal lateral;
supratemporal process of squamosal indistinct;
skull roof rugose; angular process of dentary well
developed; transition to tricuspid crowns at Tooth
10 or 11 on dentary; snout scale pattern Type II (3
rows between internasals and anterior frontals,
with posterior row composed of 3 scales in con-

SYSTEMATICS OF THE LIZARD GENUS LEJOCEPHALUS 33

tact with anterior frontals); parietal scale pattern
Type II (most lateral parietal scales subequal to
or slightly larger than median pair, and single row
[occasionally 2] of postparietals); 3 internasals,
usually 2 in contact with rostral; 4 lorilabial
scales anterior to enlarged subocular; cephalic
scale ridges extending onto frontonasals; lateral
neck scales undifferentiated; lateral trunk scales
not reduced; middorsal crest moderate, 43-62
dorsal crest scales occiput—vent; postanal es-
cutcheons usually 4; tricarinate scales of first and
second toe enlarged, fringe moderate; lateral neck
folds moderate; nuchal fold moderately convex;
lateral fold on trunk absent; throat heavily pat-
terned with chevrons or smudges; dorsum lineate.

Subspecies.—stictigaster Schwartz (1959b);
astictus Schwartz (1959b); celeustes Schwartz
and Garrido (1968a); exotheotus Schwartz
(1959b); gibarensis Schwartz and Garrido
(1968a); lipomator Schwartz and Garrido (1968a);
lucianus Schwartz (1960a); naranjoi Schwartz
and Garrido (1968a); ophiplacodes Schwartz
(1964); parasphex Schwartz (1964); septen-
trionalis Garrido (1975); sierrae Schwartz
Gi959b):

Remarks.—Formerly confused with
Leiocephalus cubensis, L. stictigaster was made
specifically distinct by Schwartz (1959b), who
subsequently designated trinomials for numer-
ous populations.

Leiocephalus vinculum Cochran, 1928b

Leiocephalus personatus vinculum—Cochran,
1941.
Leiocephalus vinculum—Schwartz, 1967a.

Holotype.—MCZ 25435. Type locality: Pointe
a Raquettes, Ile de la Gonave, Haiti.

Distribution.—lIle de la Gonave, Haiti.

Characterization.—Moderate size (males to
77 mm, females to 73 mm SVL); nasal overlap of
premaxillary spine complete; nasal-maxillary
suture curved; nasal processes of frontal not ex-
posed; frontal narrow; parietal table U-shaped in
adults; supratemporal ventromedial; supratempor-
al process of squamosal indistinct; skull roof
smooth; angular process of dentary well devel-
oped; transition to tricuspid crowns at Tooth 10
or 11 on dentary; snout scale pattern Type III (2
rows of scales between internasals and anterior
pair of frontals, with posterior row composed of
3, or a single pair of, enlarged scales); parietal
scale pattern Type III (most lateral parietals larger
than median pair, postparietals few or absent); 3
internasals, usually 2 in contact with rostral; 4
lorilabial scales anterior to enlarged subocular;
cephalic scales essentially smooth; lateral neck
scales undifferentiated; lateral trunk scales not
reduced; middorsal crest moderate; 51—60 mid-
dorsal crest scales occiput—vent; postanal es-
cutcheons usually 4; tricarinate scales of first and
second toe enlarged, fringe moderate; lateral neck
folds simple; nuchal fold moderately convex:
lateral fold on trunk absent; scapular or
suprascapular pattern absent.

Remarks.—With the reallocation of
Leiocephalus altavelensis (Noble and Hassler,
1933) to L. barahonensis, and the recognition of
L. endomychus (Schwartz, 1967a) as a distinct
species, L. vinculum 1s restricted to the mono-
typic population on Ile de la Gonave.

SUMMARY

The iguanian lizard genus Leiocephalus is
endemic to the West Indies, where 21 extant
species are found on Cuba, the Bahamas, and
Hispaniola. Two species that became extinct his-
torically and six others known only by fossils
document a previously greater range in the
Antilles.

All of the species are terrestrial, largely xe-
rophilous, and range from 55 mm to nearly 140
mm SVL; some fossil species may have reached
200 mm SVL. Living Leiocephalus resemble

other spiny, scansorial iguanians in general habitus
and none is especially apomorphic. The mono-
phyly of the genus is supported by at least 11
morphological synapomorphies. Species of
Leiocephalus are members of a group that is
otherwise exclusively South American—the
Tropiduridae. My analysis of skeletal, integu-
mentary, and soft-anatomical characters corrobo-
rates other recent studies that place them in the
topology:[Liolaeminae [Leiocephalus, Tropiduri-
nae|].

54 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Relationships within Leiocephalus also were
estimated with morphological criteria. Numer-
ous potential character transformations were iden-
tified and a restricted suite that met the demands
of discrete variation was chosen. Of the approxi-
mately 140 potential characters, 39 proved useful
for estimating a phylogeny of the species. Based
on these data, 12 equally parsimonious tree to-
pologies were obtained. These trees differ in the
placement of several terminal branches within
three primary stems, which are as follow: (1) a
clade composed of L. carinatus, greenwayi, and
punctatus; (2) a complex here referred to as the L.
melanochlorus group—melanochlorus , psammo-
dromus, inaguae, schreibersi, and macropus ; and
(3) all remaining extant species except herminieri
and eremitus. This large branch of 11 species
configures as (((loxogrammus + raviceps) +
(lunatus + personatus)) + (cubensis + Sticti-
gaster)) + (rhutidira + semilineatus) + (vincu-
lum) + (barahonensis). There are equally parsi-
monious arrangements within this scheme for
lunatus (as the sister species of loxogrammus +
raviceps), and for barahonensis with respect to
rhutidira, semilineatus, and vinculum. Leioce-
phalus pratensis displays no apparent affinity
with any one species or group of species, and
falls out as the sister species of this large, mostly
Hispaniolan complex. My analysis reveals that L.
vinculum, heretofore composed of three subspe-
cific populations, is restricted to the nominate
population on Ile de la Gonave. Leiocephalus
vinculum altavelensis is placed in the synonymy
of L. barahonensis, whereas L. v. endomychus is
recognized as specifically distinct and the sister
species of L. rhutidira.

The problematic Leiocephalus herminieri,
presumed to have come from Martinique but now
extinct and known by only four specimens, pos-
sesses a peculiar combination of primitive and
derived attributes that places it as the sister spe-
cies of all other Leiocephalus. The best candidate
for the least apomorphic extant species is L.
carinatus. The status of L. eremitus, the extinct
species of Navassa Island, remains ambiguous
because only the holotype is known and most of
its osteological characters could not be scored.

Possible relationships based on external charac-
teristics are with L. greenwayi, punctatus, and
carinatus, or with the Leiocephalus melano-
chlorus group.

None of the six named fossil Leiocephalus
from the West Indies can be placed confidently on
the tree of extant species because they are repre-
sented only by isolated and usually incomplete
skeletal elements. Of the 13 cranial characters
used in the primary analysis, none of these fossil
taxa could be scored for more than eight of them.
Inclusion of these six species into the primary
data matrix effects tree topologies of the extant
species mainly in the form of polytomies at ter-
minal stems. Some of the fossil species, for ex-
ample L. anonymous and L. apertosulcus, in fact
may be more closely related to one another than
to any extant species, based on an open Meckel’s
groove as a derived reversal. In any event, all of
these West Indian fossil forms are clearly
Leiocephalus, in contrast to others from Oligo-
cene and Miocene deposits of North America.
Most of these latter fossils are simply too incom-
plete for confident identification; their referral to
Leiocephalus was based on possession of dentaries
withaclosed Meckel’s groove and tricuspid tooth
crowns posteriorly, and neither of these charac-
ters is compelling evidence for relationship. The
most abundant material comes from the Val-
entine Formation of Nebraska and was de-
scribed as Leiocephalus nebraskensis, but
this form is neither Leiocephalus nor likely
even a tropidurid.

Leiocephalus is relictual in the West Indies.
Several of the fossil species demonstrate that the
clade’s former range included all of the major
islands in the Greater Antilles and probably most
banks of the Lesser Antilles, at least south to
Martinique. New fossils from Anguilla and
Guadeloupe are described herein and tentatively
referred to L. cuneus, a fossil species known
previously from Barbuda and Antigua. The causes
of the extinction of these Leiocephalus may or
may not be the same; where accurate chronolo-
gies are available, their demise is synchronous
with European and African settlement of the
Antilles.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 55

LITERATURE CITED

ALEXANDER, S. A., AND P. F.A. MADERSON. 1972. Fur-
ther observations on holocrine epidermal special-
izations in iguanid lizards. Am. Zool. 12(4):731—
32"

ArnoLp, E. N. 1985. Variation in the cloacal and
hemipenial muscles of lizards and its bearing on
their relationship. Symp. Zool. Soc. London 53:47—
85.

Armas, L. F. DE. 1987. Notas sobre la alimentacion de
Leiocephalus carinatus cayensis (Sauria:
Iguanidae). Poeyana 350.

BouLenceR, G. A. 1885. Catalogue of the lizards in the
British Museum (Natural History). London: British
Museum (Natural History).

Barsour, T. 1914. A contribution to the zoogeography
of the West Indies, with especial reference to am-
phibians and reptiles. Mem. Mus. Comp. Zool.
Harvard Univ. 44(2):209-359.

Barzour, T. 1915. Recent notes regarding West Indian
reptiles and amphibians. Proc. Biol. Soc. Washing-
ton 27:71-78.

Barsour, T. 1916a. Additional notes on West Indian
reptiles and amphibians. Proc. Biol. Soc. Washing-
ton 29:215-—220.

Barzour, T. 1916b. The reptiles and amphibians of the
Isle of Pines. Ann. Carnegie Mus. 10(2):297—308.

Barsour, T. 1920. A Leiocephalus misnamed. Copeia
(1920)85:73.

Barsour, T. 1935. A second list of Antillean reptiles
and amphibians. Zoologica 19(3):77-141.

Barsour, T., AND B. SHREVE. 1935. Concerning some
Bahamian reptiles,with notes on the fauna. Proc.
Boston Soc. Nat. Hist. 5:347—365.

BARBOUR, T., AND G. K. NosB-e. 1915. A revision of the
lizards of the genus Ameiva. Bull. Mus. Comp.
Zool. Harvard Univ. 59(6):416-479.

BAsKIN, J. N., AND E. E. WILLIAMS. 1966. The Lesser
Antillean Ameiva. Stud. Fauna Curagao Carib. Ilds.
89:143-176.

Bryant, H. N. 1989. An evaluation of cladistic and
character analyses as hypothetico-deductive proce-
dures, and the consequences of character weight-
ing. Syst. Zool. 38(3):214—227.

CLoucH, G. C., AND G. PuLk. 1971. The vertebrate
fauna and the vegetation of East Plana Cay, Bahama
Islands. Atoll Res. Bull. 138:1—17.

Cocuran, D. M. 1928a. A new genus and species of
lizard, Hispaniolus pratensis, from the Haitian
Republic. Proc. Biol. Soc. Washington 41:49-52.

Cocuran, D. M. 1928b. The herpetological collections
made in Haiti and its adjoining islands by Walter J.
Eyerdam. Proc. Biol. Soc. Washington 41:53—59.

Cocuran, D. M. 1931. New Bahamian reptiles. J.
Washington Acad. Sci. 21(3):39-40.

Cocuran, D. M. 1932. Two new subspecies of lizards
of the genus Leiocephalus from Hispaniola. Proc.
Biol.. Soc. Washington 45:177-182.

Cocuran, D. M. 1934a. A new lizard, Leiocephalus
personatus lunatus, from the Dominican Republic.
Occas. Pap. Boston Soc. Nat. Hist. 8:153—156.

Cocuran, D. M. 1934b. Herpetological collections
made in Hispaniola by the Utowana expedition,
1934. Occas. Pap. Boston Soc. Nat. Hist. 8:163-188.

Cocuran, D. M. 1934c. Herpetological collections
from the West Indies made by Dr. Paul Bartsch
under the Walter Rathbone Bacon Scholarship 1928—
1930. Smithson. Misc. Coll. 92(7):1-48.

Cocuran, D. M. 1941. The Herpetology of Hispaniola.
Bull. U.S. Natl. Mus. 177.

Cocnran, D. M. 1961. Type Specimens of Reptiles and
Amphibians in the U.S. National Museum. Bull.
U.S. Nat. Mus. 220.

Cope, E. D. 1863. Contributions to neotropical
saurology. Proc. Acad. Nat. Sci. Philadelphia
14:176-188.

Cope, E. D. 1868. An examination of the Reptilia and
Batrachia obtained by the Orton expedition to
Equador and the upper Amazon, with notes on other
species. Proc. Acad. Nat. Sci. Philadelphia 20:96—
123:

Cope, E. D. 1887. List of the Batrachia and Reptilia of
the Bahama Islands. Proc. U.S. Natl. Mus. 10:436—
439.

De Queiroz, K. 1987. Phylogenetic Systematics of
I guanine lizards,a Comparative Osteological Study.
Univ. California Berkeley Publ. Zool. 118.

DumeriL, A. M. C, AND G. Brpron. 1837. Erpétologie
Générale ou Histoire Naturelle Complete des Rep-
tiles. Vol. 4. Paris: Roret.

Dunn, E. R. 1920. A new lizard from Haiti. Proc. New
England Zool. Club. 7:33-34.

Estes, R. 1963. Early Miocene salamanders and liz-
ards from Florida. Q. J. Florida Acad. Sci.
26(3):234—256.

Estes, R. 1983. Handbuch der Paldoherpetologie.
Encyclopedia of paleoherpetology. Part 1OA, Sauria
terrestria, Amphisbaenia. Stuttgart and New York:
Gustav Fischer.

Estes, R., J. GAUTHIER, AND K. DE QuetRoz. 1988.
Phylogenetic relationships within Squamata.. Pp
119-281 inR. Estes and G. Pregill (eds.), Phyloge-
netic Relationships of the Lizard Families, Essays
Commemorating Charles L. Camp. Stanford, Cali-
fornia: Stanford Univ. Press.

56 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Estes, R., AND J. A. TIHEN. 1964. Lower vertebrates
from the Valentine Formation of Nebraska. Am.
Midl. Nat. 72(2):453-472.

ETHERIDGE, R. 1964. Late Pleistocene Lizards from
Barbuda, British West Indies. Bull. Florida State
Mus. Biol. Ser. 9(2):43—75.

ETHERIDGE, R. 1965. Fossil lizards from the Dominican
Republic. Q. J. Florida Acad. Sci. 28(1):83-105.

ETHERIDGE, R. 1966a. The systematic relationships of
West Indian and South American lizards referred to
the iguanid genus Leiocephalus. Copeia
1966(1):79-91.

ErueripGe, R. 1966b. An extinct lizard of the genus
Leiocephalus from Jamaica. Q. J. Florida Acad.
Sci. 29(1):47-59.

ETHERIDGE, R. Anew psammophilus lizard of the genus
Liolaemus (Squamata, Iguania, Tropiduridae) from
northwestern Argentina. Boll. Mus. Region Sci.
Nat. Torino. In press.

ETHERIDGE, R. AND K. DE Queiroz. 1988. A phylogeny
of Iguanidae. Pp 283-367 in R. Estes and G. Pregill
(eds.), Phylogenetic Relationships of the Lizard
Families, Essays Commemorating Charles L. Camp.
Stanford, California: Stanford University Press.

Evans, L. T. 1953. Tail display in an iguanid lizard.
Copeia 1953(1):50-54.

Farris, S. 1988. Hennig86 (Ver. 1.5) Reference.

Firzincer, L. I. 1826. Neue Classification der Reptilien
nach ihren Nattirlichen Verwandtschaften nebst
einer Verwandts-Tafel und einem Verzeichnisse der
Reptilien-Sammlung des k.k. Zoologischen Mu-
seum zu Wien. Wien: J. G. Hiner.

Fitzincer, L. I. 1843. Systema Reptilium. Fasciculus
primus. Wien: Baumiiller and Seidel.

Frost, D. R. 1987. A phylogenetic analysis of the
Tropidurus Group of I[guanian Lizards, with Com-
ments on the Relationships Within the Iguania
(Squamata). Ph.D. Dissertation. Lawrence: Univ.
Kansas.

Frost, D. R., AND R. ETHERIDGE. 1989. A phylogenetic
analysis and taxonomy of iguanian lizards (Reptilia:
Squamata). Univ. Kansas Mus. Nat. Hist. Misc.
Publ. 81:1—65.

GALI, F., AND A. SCHWARTZ. 1982. A new subspecies of
Leiocephalus personatus from the Republica
Dominicana. J. Herpetol. 16(2):177—179.

Gaul, F., A. SCHWARTZ, AND A. SUAREZ. 1988. A new
subspecies of Leiocephalus personatus (Sauria:
Iguanidae) from Haiti. Proc. Biol. Soc. Washington
101(1):1-3.

GARMAN, S. 1887. On the reptiles and batrachians of
Grand Cayman. Proc. Am. Philos. Soc.
24(126):273-286.

Garripo, O. H. 1973a. Nueva especie de Leiocephalus

(Lacertilia, Iguanidae) para Cuba. Poeyana 116:1—
I)

Garripo, O. H. 1973b. Nuevas subespecies de reptiles
para Cuba. Torreia, n.s. 30:1—28.

Garripo, O. H. 1975. Nueva reptiles del archipiélago
Cubano. Poeyana 141:1—58.

Garripo, O. H. 1979. Nuevas subespecies de
Leiocephalus macropus Cope (Lacertilia:
Iguanidae) para Cuba. Poeyana 188:1—16.

Gorman, G. C., L. ATKINS, AND T. HoLzINGER. 1967.
New karyotypic data on 15 genera of lizards in the
family Iguanidae, with a discussion of taxonomic
and cytological implications. Cytogenetics 6:286—
299:

Gravenuorst, J. C. L. 1837. Beitraége zur genauern
Kenntniss einger Eidechsengattungen. Nova Acta
Acad. Leop. Carol. Halle 18(2):712—784.

Gray, J. E. 1827. A description of a new genus and
some new species of saurian reptiles; with a revi-
sion of the species of chameleons. Philos. Mag.
2(2):207-214.

Gray, J. E. 1840. Catalogue of the species of reptiles
collected in Cuba by W. S. MacLeay, Esq; with
some notes on their habits extracted from his MS.
Ann. Mag. Nat. Hist. 5:108—115.

Gray, J. E. 1845. Catalogue of the specimens of lizards
in the collection of the British Museum. London:
Taylor and Francis.

Gunbacn, J. 1880. Contribucion a la Erpetologia
Cubana. La Habana: Montiel.

HaLLoweELL, E. 1856. Notes on the reptiles in the
collection of the Museum of the Academy of Natu-
ral Sciences. Proc. Acad. Nat. Sci. Philadelphia
7:146-153.

Harpy, J. D. 1958a. A new lizard of the genus
Leiocephalus from Cuba (Squamata: Iguanidae). J.
Washington Acad. Sci. 48(9):294—300.

Harpy, J. D. 1958b. A geographic variant gradient in
the Cuban lizard, Leiocephalus macropus Cope.
Herpetologica 13:275-276.

HENNIG, W. 1966. Phylogenetic Systematics. Urbana:
Univ. Illinois Press.

Ho.mon, J. A., AND R. M. SuLtivan. 1981. A small
herpetofauna from the type section of the Valentine
Formation (Miocene: Barstovian), Cherry County,
Nebraska. J. Paleontol. 55(1):138—144.

KuinGeL, G. C. 1932. Shipwrecked on Inagua. Nat.
Hist., 33:42-55.

KLINGEL, G. C. 1941. The Ocean Island (Inagua). New
York: Dodd Mead.

Kiuce, A. G. 1989. A concern for evidence and a
phylogenetic hypothesis of relationship among
Epicrates. (Boidae, Serpentes). Syst. Zool. 38(1):7—

ZS.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS B/

Lana, M. 1989. Phylogenetic and biogeographic pat-
terns of Basiliscine Iguanians. Zool. Monogr. Bonn
28:1-172.

Mappison, W. D., M. J. DONOGHUE, AND D. R. MADDISON.
1984. Outgroup analysis and parsimony. Syst. Zool.
33(1):83-103.

Marce ini, D. L., AND T. A. JENSSEN. 1989. Thermal
ecology of the tropical iguanid lizard, Leiocephalus
schreibersi. Am. Midl. Nat. 122:44—50.

MertENS, R. 1939. Herpetologische Ergebnisse einer
Reise nach der Insel Hispaniola, Westindien. Abh.
Senekenber. Naturf. Ges. 449:1—84.

Mirera, J. F. 1984. Agresividad de Leiocephalus
cubensis Gray 1840 (Reptilia: Sauria: Iguanidae).
Misc. Zool. 22:2.

Nose, G. K. 1923. Four new lizards from Beata
Island, Dominican Republic. Am. Mus. Novit. 64:1—
3

Nose, G. K., AND W. G. HAss.er. 1933. Two new
species of frogs, five new species and a new race of
lizards from the Dominican Republic. Am. Mus.
Novit. 652:1—-17.

NosLe, G. K., AND G. C. KLINGEL. 1932. The reptiles of
Great Inagua Island, British West Indies. Am. Mus.
Novit. 549:1—25.

Nore, M.A. 1989. Late Cenozoic lizards of the Anza
Borrego Desert, California. Nat. Hist. Mus. Los
Angeles Cty. Conrib. Sci. 414:1—31.

Oxson, S. L., G. K. PREGILL, AND W. B. HILGARTNER.
1990. Studies on fossil and extant vertebrates from
San Salvador (Watling’s) Island, Bahamas.
Smithson. Contrib. Zool. 508:1—15.

PAULL, D., E. E. WILLIAMS, AND W. P. HA tv. 1976.
Lizard karyotypes from the Galapagos Islands:
chromosomes in phylogeny and evolution. Breviora
441:1—31.

Porter, C. A., R. I. CRomBig, AND R. J. BAKER. 1989.
Karyotypes of five species of Cuban lizards. Occas.
Pap. Mus. Texas Tech Univ. 130:1-6.

PREGILL, G. 1981. Late Pleistocene herpetofaunas from
Puerto Rico. Univ. Kansas Mus. Nat. Hist. Misc.
Publ. 71:1—72.

PREGILL, G. K. 1984. Anextinct species of Leiocephalus
from Haiti (Sauria: Iguanidae). Proc. Biol. Soc.
Washington 97(4):827-833.

PREGILL, G. K. 1986. Body size of insular lizards: a
pattern of Holocene dwarfism. Evolution 40(5):997—
1008.

PREGILL, G. K., AND S. L. OLSON. 1981. Zoogeography
of West Indian vertebrates in relation to Pleistocene
climatic cycles. Ann Rev. Ecol. Syst. 12:75—98.

PREGILL, G. K., AND D. W. STEADMAN. Late Quaternary
vertebrate localities of the Lesser Antilles. Manu-
script.

PREGILL, G. K., D. W. STEADMAN, S. L. OLSON, AND F. V.
Graby. 1988. Late Holocene fossil vertebrates from
Burma Quarry, Antigua, Lesser Antilles. Smithson.
Contrib. Zool. 463:1—27.

PREGILL, G. K., R. I. CRomBire, D. W. STEADMAN, L. K.
Gorpon, F. Davis, AND W. B. HILGARTNER. 1991.
Living and late Holocene fossil vertebrates and the
vegetation of the Cockpit Country, Jamaica. Atoll
Res. Bull. No. 353:1-18.

Rass, G. B. 1957. A study of variation in iguanid
lizards of the Leiocephalus carinatus complex.
Ph.D. Dissertation. Ann Arbor: Univ. Michigan.

RosInson, M. D., AND T. R. VAN DEVENDER. 1973.
Miocene lizards from Wyoming and Nebraska.
Copeia 1973(4):698-704.

Rosinson, P. L. 1967. The evolution of the Lacertilia.
Collog. Int. C. N. R. S. 104:243-279.

Ropricues, M. T. 1987. Sistematica, ecologia e
zoografia dos Tropidurus do grupo torquatus ao Sul
do Rio Amazonas. Arq. Zool. Sao Paulo 31(3):105—
230.

SAGRA, R. D. DE LA 1837. Historia Fisca Politica y
Natural la Isla de Cuba I. Reptiles y Peces. Paris:
A. Bertrand.

ScuMipT, K. P. 1921a. Notes on the herpetology of
Santo Domingo. Bull. Am. Mus. Nat. Hist. 44(2):7—
20.

ScuMipT, K. P. 1921b. The herpetology of Navassa
Island. Bull. Am. Mus. Nat. Hist. 44(18):555—559.

ScHMIpT, K. P. 1936. Notes on Bahaman reptiles and
amphibians. Field Mus. Nat. Hist. Zool Ser.
20(16):127-133.

SCHOENER, T. W., J. B. SLADE, AND C. H. STINSON. 1982.
Diet and sexual dimorphism in the very catholic
lizard genus Leiocephalus of the Bahamas, West
Indies. Oecologia Berlin 52(2):160—-169.

ScHwartz, A. 1959a. The Cuban lizards of the species
Leiocephalus carinatus (Gray). Reading Univ. Pub.
Mus. Art Gal. Sci. Publ. 10:1-47.

ScHwartz, A. 1959b. Variation in lizards of the
Leiocephalus cubensis complex in Cuba and the
Isla de Pinos. Bull. Florida State Mus. Biol. Ser.
4(4):97-143.

ScHwartz, A. 1960a. Anew subspecies of Leiocephalus
stictigaster Schwartz from central Cuba. Proc. Biol.
Soc. Washington 73:103—106.

ScHwartz, A. 1960b. Variation in the Cuban lizard
Leiocephalus raviceps Cope. Proc. Biol. Soc. Wash-
ington 73:67—82.

Scuwartz, A. 1964. New subspecies of Leiocephalus
from Cuba. Q. J. Florida Acad. Sci. 27(3):211-—
DD:

Scuwartz, A. 1965. The Leiocephalus (Lacertilia,
Iguanidae) of Hispaniola I. Leiocephalus

58 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

melanochlorus Cope. J. Ohio Herpetol. Soc.
5(2):39-48.

Scuwartz, A. 1967a. The Leiocephalus (Lacertilia,
Iguanidae) of Hispaniola, II. The Leiocephalus
personatus complex. Tulane Stud. Zool. 14(1):1-53.

Scuwartz, A. 1967b. The Leiocephalus (Lacertilia,
Iguanidae) of the southern Bahama Islands. Ann.
Carnegie Mus. 39(12):153-185.

Scuwartz, A. 1968. The Leiocephalus (Lacertilia,
Iguanidae) of Hispaniola. III. Leiocephalus
schreibersi, L. semilineatus, and L. pratensis. J.
Herpetol. 1(1—-4):39-63.

ScHwartz, A. 1969. Two new subspecies of
Leiocephalus from Hispaniola. J. Herpetol. 3(1—
2):79-85.

Scuwartz, A. 1971. Anew subspecies of Leiocephalus
personatus (Sauria: I[guanidae). Herpetologica
27(2):176-182.

Scuwartz, A. 1979a. A new species of Leiocephalus
(Reptilia: Iguanidae) from Hispaniola. Proc. Biol.
Soc. Washington 92(2):272-279.

ScHwartz, A. 1979b. The herpetofauna of Ile a Cabrit,
Haiti, with the description of two new subspecies.
Herpetologica 35(3):248—255.

ScHWARTZ, A., AND O. H. Garripo. 1967. A review of
the Cuban iguanid lizard Leiocephalus macropus
Cope. Reading Univ. Pub. Mus. Art Gal. Sci. Publ.
14:1-41.

ScHwartz, A., AND O. H. GArRripo. 1968a. Four new
subspecies of Leiocephalus stictigaster from Cuba.
Nat. Mus. Canada Nat. Hist. Pap. 37:1—23.

SCHWARTZ, A., AND O. H. Garripo. 1968b. An
undescribed subspecies of Leiocephalus raviceps
Cope (Sauria: Iguanidae) from western Cuba. Proc.
Biol. Soc. Washington 81:23-30.

ScHwartTz, A., AND R. W. HENDERSON. 1988. West In-
dian amphibians and reptiles: a check-list. Contrib.
Biol. Geol., Milwaukee Pub. Mus. 74:1—264.

Scuwartz, A., AND L. H. OGREN. 1956. A collection of
reptiles and amphibians from Cuba, with the de-
scription of two new forms. Herpetologica 12(2):91—
110.

ScHwartz, A., AND R. THOMAS. 1975. A check-list of
West Indian amphibians and reptiles. Spec. Publ.
Carnegie Mus. Nat. Hist. 1:1—216.

SCHWARTZ, A., R. THOMAS, AND L. D. OBer. 1978. First
Supplement to a check-list of West Indian amphib-
ians and reptiles. Spec. Publ. Carnegie Mus. Nat.
Hist. 5:1-35.

SeToGcucul, T. 1978. Paleontology and geology of the
Badwater Creek Area, central Wyoming. Part 16.
The Cedar Ridge Local Fauna (late Oligocene).
Bull. Carnegie Mus. Nat. Hist. 137:379-436.

Situ, H. M. 1946. Handbook of Lizards. New York:
Comstock Publ. Co.

STEADMAN, D. W., PREGILL, G. K., AND S. L. OLSON.
1984. Fossil vertebrates from Antigua, Lesser
Antilles: evidence for late Holocene human-caused
extinctions in the West Indies. Proc. Natl. Acad.
Sci. 81:4448-445 1.

STEJNEGER, L. 1901. Diagnosis of a new species of
iguanoid lizard from Green Cay, Bahama Islands.
Proc. U.S. Natl. Mus. 23(1219):471.

STEJNEGER, L. 1904. The herpetology of Porto Rico.
Rept. U.S. Natl. Mus. for 1902, 129:549-724.
STEJNEGER, L. 1917. Cuban amphibians and reptiles
collected for the United States National Museum
from 1899-1902. Proc. U.S. Natl. Mus. 53:259-

291%

Sworrorb, D. L. 1985. PAUP—Phylogenetic Analysis
Using Parsimony. Version 2.4.

Tuomas, R. 1966. A reassessment of the herpetofauna
of Navassa Island. J. Ohio Herpetol. Soc. 5(3):73—
89.

Tscuupt, J. J. von. 1845. Reptilium conspectus quae in
Republica Peruana reperiuntur et pleraque observata
vel collecta sunt in itinere a Dr. J. D. de Tschudi.
Arch. Naturgesch. 11(1):150—170.

Varona, L. S., AND O. H. Garripo. 1970. Vertebrates
of Cayos de San Felipe, Cuba, including a new
species of hutia. Poeyana 75:—126.

Watters, D.L., E. J. Reitz, D. W. STEADMAN, AND G. K.
PREGILL. 1984. Vertebrates from archaeological sites
on Barbuda, West Indies. Ann. Carnegie Mus. Nat.
Hist. 53(13):383-412.

WELLSTEAD, C. F, 1982. Lizards from the lower Valen-
tine Formation (Miocene) of northern Nebraska. J.
Herpetol. 16(4):364—375.

WELLSTEAD, C. F. 1983. Leiocephalus nebraskensis
nom. nov. pro L. septentrionalis Wellstead, 1982, a
junior homonym. J. Herpetol. 17(4):408.

Witey, E. O. 1981. Phylogenetics. The Theory and
Practice of Phylogenetic Systematics. New York:
John Wiley, Sons.

WILuiAMs, E. E. 1963. Anolis whiteman, new species
from Hispaniola (Sauria, Iguanidae). Breviora
197:1-8.

WILLIAMS, E. E. 1 988. A new look at the I[guania. Pp.
429-488 in W.R. Heyer and P. E. Vanzolini (eds.),
Proceedings of a Workshop on Neotropical Distri-
bution Patterns. Rio de Janeiro: Acad. Brasil Cien.

WILLIAMS, E. E. 1989. Old problems and new opportu-
nities in West Indian biogeography. Pp 1-46 in C.
A. Woods (ed.). Biogeography of the West Indies:
Past, Present, and Future. Gainesville, Florida:
Sandhill Crane Press.

ZuG, G. R. 1959. Three new subspecies of the lizard
Leiocephalus macropus Cope from Cuba. Proc.
Biol. Soc. Washington 72:139-150.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS

SQ OS Se) SS

SSS

0

0

fo)

0

oo)

0

foe)

OG OC 5 OOO OOOO TO GOS OOO FO Os O00 Om O74 ~ WNINIUIA
LOC l Ob OO OO GO LOCO COO Tl LOO OOO OO WO 1 4IISDSNIUS
Oy GO O41 OO UO OW 6 OFS OO OV PO YO Wo 1 O-O SHIDIUI] US
LOO te Ok OI OL 0.6 OF WO Wr OO OO) OO On 1 ISAIGIAAY IS
OOCOTO CO OOOO OO CO SOW Pek OO OO Ww oo DAIPUNY.t
Ct CO LOM OWL ee tl eos Oy OW tO OO Ore Oni SdaJ1A.t
OG OT OOO fF OL OO OLS COON OW Wa 0 tO OO Oo snypjound
OM SO OO CeO OO ee OT INO OOO Oh ie OF 1 1 o sniuojpoununsd
COW OC Oe OOM WO? OS OSLO OOO WOOO C1 OO OO 7 sisuajpad
ONO 6G ON SOOO OO WO eC OO iO OL OO OOo 1 snjpouos.ad
OMe 07000 OI 1) OF08G 1 0 0 00) Or 000). OO hl OMmOnOn Onl Sno] YIOUD] aU
Oe! OO, OF TOM eG Oe. OW 2 OOOO 2 OO FOO OF i sndo.gpu
OO COT Oi OO OOO Wy OIE a0) Cae OO) Wh ONO: OO Oot OO 1 snypun]
CUCOTO OL OOO OT Gis kOe Oe ek oO 1 fh OLO OO SNUUDASOXO]
MS OLONO 1 EEO) SOS OF OE ON AO) OOOO) Oe th OLOrO L oO aDNSDUI
OOS 10F OF ORO 0) O20 Os0N0 0 00 Sh Os 00 OF OMI AOROMO SON l e201 OOO LO 1UNULOY
OmOmen Om 0 10 se La LOnO LOMO Tt On c= lh CaOmOnOnOn I OmOn Tl 1On OM lKOmOmOnmO IADMUIIAS
OPO GO Oral 6 On 0) 1, OOr Oe 1) Ny Orne Ourlas ren Or eGaee alec ecn een reece cee SnjMuasa
OCCT OW OO WoO ie eNO Ol TOO ONO OO OI sIsuaqny
OOCOU OOM OOOO O Ks eo OOe OOO OO £ OO W SNIDULIDI
OOM OG eT ONO OO OW OS OE OO OF O10 OOO. 0.0. O 1 SISUdUOYD.DG
OPOMOMOMO CONG Go Gee 0) 001010) O10) O10! O00) 0000 ONO 10N 0) a0) 0 0m0 Joysoouy
Soe AC. le OmGris: on Om Gastar OC-o1. G0 e608 Le On Sofa C mlatO moma OF Coat: Gara aPON
Cede BU Mt eae AG AC © ICOM AG tGs OG Cael i eek ube Me ae eels Bical
TOJOVIV YD)

(81 ‘OI-Ppl “SSI JO UOTON.QSUOD 991) UL posn) sUOTRUIOJSUR.D JO}DRILYO SsNj]MYdaI01dT JOJ XLeU eVep Indu]

‘T XIGNUddV

60 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

APPENDIX II.

Apomorphy list for tree in Figure 14; based on data matrix from Appendix I.

Stem Character Ancestral state Derived state

barahonensis 23 0) 1
30 0 D

carinatus Ds 1 0
cubensis 38 0 1
eremitus 2A 0 1
greenwayl iM 0
26 0 1

28 0 1

36 0) 1

herminieri 6 @) 1
18 0 1

18 3 1

2 1 0

39 0) 1

inaguae 6 l 0
10 1 0

16 1 2

21 1 0)

35 0 l

loxogrammus ] l 0
5) 0 1

6 0 1

8 0 l

11 1 0

lunatus 4 0 ]
12 | 0

36 0) l

macropus 5) 0 l
20 l 2

29) | 2

36 0 l

38 0 2.

melanochlorus p} l 0)
20 | p)

22 | 2

29 l 0

personatus 19 I 0

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 61

Appendix II Continued

Stem Character Ancestral state Derived state

38 0 l
pratensis 20 l 2
Di, l 2
30 0 l
38 0 2
psammodromus 9 0 1
punctatus 2 0
Ti 0
| 0
raviceps DD l Z
28 0 |
34 0 l
rhutidira 3) 0 1
3 0
29 1 2
schreibersi 7 1 0
12 0 l
14 0 1
28 0 |
35 0 1
semilineatus 9 0 1
1) 0 ]
38 0 2
Stictigaster
vinculum 11 1 0
29 1 0
] 16 3 2
DD, l 2
2 14 0 l
24 0 ]
38 0 2
3 27h | 0
4 15 0
19 0 |
5 2D 0 l

62 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Appendix II Continued

Stem Character Ancestral state Derived state

6 4 0
23 0)

i] 9 1 0

Sil 2 1

8 18 3 D

33 1 2

9 a 1 0

0) 0 1

at 0 1

12 0 1

22 | 0

10 16 1 3

25 1 0

Bil 0 2

33 ) |

1] 1 1 @)

3 0) l

8 0 1

22 1 0

28 0 l

12 6 0 I

10 0 1

21 0 |

37 0 |

39 0) |

13 12 | 2

52 0 l

14 2 0 l

18 8 |

26 0) |

24 0) l

15 8 0) l
l

eS)
=
S

16 16

ws)
Wey Cnet

in)

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS

Appendix II Continued

63

Derived state

Stem Character Ancestral state

18 1 0 |
5) 1 0)
7 0 l
oD 0 1
29 Ol

19 20 0) |

20 16 0 1
18 0 3
30 0) 3

APPENDIX III.

List of changes within transformation series for tree in Figure 14. U = unpolarized, UO =

unordered transformation.

Changed
Character From To Along stem Consistency

1 0 1 18

i 0 11

I 0 loxogrammus 0.333
2 0 1 14

0 1 punctatus

1 0 melanochlorus 01333
3 0 i 11 1.000
4 0 1 6

0 1 lunatus 0.500

5U 1 0 18

0 1 rhutidira

0 1 macropus

0 1 loxogrammus 0.250
6 0 1 12

0 i loxogrammus

] 0 inaguae

64 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Appendix III Continued

Changed
Character From To Along stem Consistency
0 | herminieri 0.250
h 0) 1 18
| 0 9
1 0 schreibersi
0 1 punctatus 0.250
8 0 1 fal
0 1 15
0 1 loxogrammus 0.333
9 0 1 9
l 0 7
0 1 semilineatus
0 l psammodromus 0.250
10 0 1 ili
| 0 inaguae 0.500
11 0 1 9
1 0 vinculum
l 0 loxogrammus
0 1 herminieri
0 I greenwayi 0.200
12 @) l 9
0) 1 schreibersi
| 0 lunatus 0.333
13U 0 | rhutidira 1.000
14 0) l 2
0) | schreibersi 0.500
15 0) | 4 1.000
16U0O 0) | 20
| 3 10
l 2 13
| 2. 16
3} 2 |
l 2 inaguae 0.500

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 65
Appendix III Continued
Changed
Character From To Along stem Consistency
17 0 1 16 1.000
18U0 0 3 20
3 2 8
3 l 14
3 1 herminieri 0.750
19 0 1 4
0 1 semilineatus
] 0 personatus 0333
20 0 l 19
| 2 pratensis
1 2, melanochlorus
l pr macropus 0.500
21 0 | 12
0 1 punctatus
] 0 inaguae
0 1 eremitus 0.250
2D 0 1 18
1 0 y)
0 l 5)
1 0 11
1 2) l
1 7 raviceps
1 2 melanochlorus
1 2 macropus 0.250
23 0 | 6
0 | barahonensis 0.500
24 0 1 2 1.000
25U 1 0 10
] 0 carinatus 0.500
26U 0 1 14
0 l greenwayl 0.500
ZI UA0O 1 0 3

66

UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

Appendix III Continued

Changed
Character From To Along stem Consistency
l 2 pratensis
I 0 herminieri 0.667
28U 0 1 11
0 1 schreibersi
0 1 raviceps
0 1 greenwayi 0.250
BS) 0 l 18
| 0 vinculum
I 2 rhutidira
1 0 melanochlorus 0.500
30U0 0 i i)
1 0 15
0 1 pratensis
0 2 barahonensis 0.500
31U0 0 2 10
2 l df
0 l 16 0.667
By) 0 1 13 1.000
33U0 0 3 20
3 | 10
l 2 8
3 2 7) 0.750
34 0 l 14
0 l raviceps 0.500
35) 0 1 schreibersi
0 l inaguae 0.500
36 0 I macropus
0 l lunatus
0 | greenwayl 03333
3) 0 I 72 1.000
38U,U0O 0 2 7)

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 67

Appendix III Continued

Changed

Character From To Along stem Consistency
0 2 semilineatus
0 2 pratensis
0 l personatus
0 2 macropus
0 l cubensis 0.333

39 0 1 12

0 l herminieri 0.500

APPENDIX IV.

Specimens examined. Abbreviations are given
in the acknowledgements. S = skeleton or skeletal
series.

Leiocephalus apertosulcus: UF 10088-10101,
55802. L. anonymous: USNM(VP) 340183-
340192.

L. barahonensis: SDSNH 64667(S); USNM
80326, 80379, 224972, 224973. L. b. barahonensis:
SDSNH 64570-64574(S), 64575-64577,
64578(S), 64579(S), 64580-64583. L. b. aureus:
MCZ 68612, 68613; REE 1809(S). L. b. beatanus:
ASFS V2726, V2729, V17203; REE 1821: USNM
83875, 84283, 85059(S). L. b. oxygaster: SDSNH
64584, 64585. L. b. oxygaster X aureus: SDSNH
64586, 64587.

L. cuneus: UF 8226-8233, 8263-8271, 8444,
8468-8470; USNM(VP) 340157—340164, 340196.

L. carinatus: ASFS V22390, V22393, V22394;
MCZ 141246(S); UMMZ 149103(S); USNM
81709(S), 220646. L. c. aquarius: SDSNH 65958,
65966(S), 65983(S), 65997. L. c. armouri: ASFS
V8954, V8966, V36187; MCZ 6966(S), 6966b(S);
USNM 158895. L. c. coryi: ASFS X4746, X4747,
X4752. L. c. granti: MCZ 45127-45130; REE
1469(S). L. c. helenae: MCZ 38112, 38115. L. c.
microcyon: MCZ 11187(S), 11188(S).L.c. varius:
ASFS V11682, V11684, 16144; USNM 81750-—
81752,217299(S). L.c. virescens: MCZ 141247(S);
REE 1505.

L. cubensis: CAS 39304(S), 39305-39308:

CAS-SU 9247-9254, 14617-14623; MCZ 150341,
150342; SDSNH 65833(S), 66335(S); USNM
27999, 138643.

L. eremitus: USNM 12016.

L. etheridgei: USNM(VP) 259190-259202.

L. greenwayi: MCZ 36711, 36716(S), 162032,
162033; REE 1814(S); UMMZ 94051, 114518,
149108(S); USNM 120766, 149108(S).

L. herminieri: BM(NH) 52.12.3.10(S); MNHN
6829.

L. inaguae: ASFS 10337; LSUMZ 30265—
30270, 30272, 30275; MCZ 154263(S); UMMZ
149133(S); USNM 81277, 89375.

L. jamaicensis: AMNH 2311-2315; UF 8489—
8493, 8496, 8505, 8511-8513.

L.loxogrammus loxogrammus: MCZ 38131(S),
38135(S); SDSNH 66342(S), 66343-66345,
66346(S), 66347(S), 66348-66352, 66353(S),
66354; UMMZ 149134(S); USNM 14569. L. 1.
parnelli: KU 192293; USNM 220526, 220529,
220537, 220576—220596(S).

L. lunatus arenicolor: MCZ 75079-75082;
SDSNH 64588; USNM 40919, 40920. L. / . are-
nicolor X lewisi: SDSNH 64610-64612, 64613-
64617(S), 64618-64620, 64621(S). L. 1. lewisi:
SDSNH 64589-64599, 64600-64602(S), 64604—
64608, 64609(S); USNM RD145, RD147. L. 1.
louisie: USNM 40210, 40211. L. 1. melaenacelis:
USNM 40912, 40914, 40916. L. 1. thomasi: REE
1815(S).

68 UNIV. KANSAS MUS. NAT. HIST. MISC. PUB. No. 84

L. macropus: MCZ 11208(S); REE 1819(S). L.
m. macropus: SDSNH 65959, 65960, 65965(S),
65989(S), 66002, 66004(S), 66005(S), 66012;
USNM 25819, 220653(S), 220721. L. m. asbolo-
mus: USNM 220654—220656(S), 220722, 220723.
L. m. immaculatus: USNM 220647—220652(S).

L. melanochloris: MCZ 3598, 37533(S),
59545(S); SDSNH (4 spec. uncat.); USNM 80852,
80858(S), 80859, 80860. L. m. hypsistus: MCZ
59545.

L. nebraskensis: UNSM 47025, 47075, 47088,
47134,47144, 47146, 47148, 51812,51813,51815,
51818, 51819, 56049, 56085, 56092, 56093.

L. partitus: KUVP 11473; USNM(VP) 259203.

L. personatus: MCZ 3615; REE 1811; SDSNH
10781-10783. L. p. budeni: KU 93316-93321. L.
p.mentalis: SDSNH 64630, 64631, 64644(S). L. p.
scalaris : MCZ58038(S), 58044(S), 5805 1(S); REE
1803; USNM 224975-224978, 225044(S),
225045(S). L. p. socoensis: SDSNH 64622(S),
64623-64626, 64627(S), 64628, 64629. L. p.
tarachodes: ASFS V 16137, V16139, V16140. L. p.
trujilloensis: ASFS X9249, V14577, V14578.

L. pratensis: MCZ 61229, 56044(S); USNM
69189, 74121, 74124, 74127. L. p. chimaris: ASFS
V9841(S), V9843, V9846.

L. psammodromus: LSUMZ 30364-30368,
30373-30379, 30385; MCZ 11948: REE 1813(S);
UMMZ 149109(S); USNM 81385. L. p. aphretor:
MCZ 54191, 54192. L. p. mounax: MCZ 54170,
86141, 86143(S), 86146.

L. punctatus: ASFS V10999, V27421,
V27423(S), V27424, V27515, V27529; MCZ
38083(S), 38087(S); UMMZ 81560, 149110(S).

L. raviceps: MCZ 13376(S); UDMMZ 149111(S);
USNM 4162, 220657220665, 220724, 220729. L.
r.raviceps: SDSNH 65928, 65936, 65963,65964(S),
65985, 65986(S), 65987(S), 65993(S).

L. rhutidira: ASFS V46324, V46742,
V46748(S), V46750.

L. schreibersi: REE 1815; SDSNH 64665(S),
64666(S), 64668—64670(S), 6467 164678: USNM
40021, 40022. L. s. schreibersi: MCZ 39592(S),
59591(S), 64908, 64911, 65791(S); USNM 40021,
40022. L. s. nesomorus: MCZ 37556, 37564.

L. semilineatus: MCZ 58069, 58073; SDSNH
64632(S), 64633, 64634(S), 64635(S), 64636,
64637(S), 64638, 64639(S), 64640(S), 64641,
64642, 64643(S); USNM 40077, 40081, 225046—

225048(S), 259510(S).

L. stictigaster stictigaster: AMNH 77864; MCZ
118706, 118870. L. s. celeustes: MCZ 92021,
92022. L. s. exotheotus: MCZ 11114 5iileeoes:
gibarensis: ASFS V11763, V11764(S), V11765. L.
s. lucianus: MCZ 59228(S); REE 1810. L. s.
naranjoi: USNM 140466, 140467.

L. vinculum: MCZ 25435, 25437; REE 1812.
L. v. vinculum: ASFS X2495, V26619(S). L. v.
altavelensis: ASFS V26908, V26909. L. v. endo-
mychus: ASFS V43786, V43788, V43795.

Liolaemus: L. anomalus: MCZ 19053, 19054:
REE 2283(S). L. austromendocinus: MCZ 19110,
19112; REE 2343(S). L. bibioni: MCZ 19313,
19314; REE 2406(S).L. boulengeri: REE 2458(S).
L. darwini: MCZ 19170, 19171; REE 2495(S). L.
eleodori: REE 2376(S).L. elongatus: MCZ 19233,
19234; REE 2366(S). L. e. petrophilus: REE
2428(S). L. fitzingeri cuyanus: REE 2316(S). L. k.
kingi: REE 2481(S). L. kriegi: REE 2417(S),
2418(S). L. lineomaculatus: REE 2465(S). L.
multiformis: REE 1826(S), 1827(S). L. pictus:
REE 1897(S), 1874(S). L. rothi: REE 2398(S),
2400(S). L. ruibali: REE 2301(S).

Ophryoessoides: O. aculeatus: KU 121092,
121093(S), 121094; UMMZ 149102. O. arenarius:
WP. 544(S), 577(S). O. caducus: KU 136354,
136355; REE 2285(S). O. guentheri: WP 541(S),
549-551(S). O. iridescens: KU 121139-121141,
142683, 142695(S), 164170. O. i. cajamareae:
REE 1820(S); USNM 200912(S), 222585(S). O.
trachecephalus: REE 234(S).

Phymaturus: P. palluma: REE 2323(S),
2325(S), 2326(S). P. patagonicus patagonicus:
REE 2472(S). P. p. payuniae: REE 2336(S). P. p.
somuncurensis: MCZ 19284, 19285; REE 2436(S).

Plica: P. plica: KU 167499(S); REE 2167(S).P.
umbra: KU 125968(S); USNM 204266(S).

Procotretus: P. pectinatus: KT 187794(S),
187798(S).

Stenocercus: S. apurimacus: KU 134270,
134278(S), 134283, 134284(S), 134288. S. boett-
geri: KU 134014(S). S. chrysopygus: KU
133895(S). S. crassicaudatus: REE 2284(S),
2286(S). S. cupreus: KU 133974(S). S. empetrus:
KU 134404(S). S. festae: KU 134595(S), 141150,
141151.S.formosus: KU 134110(S).S. guentheri:
KU 147347, 169857, 192678, 192679; USNM
222584(S). S. humeralis: KU 134004(S). S.

SYSTEMATICS OF THE LIZARD GENUS LEIOCEPHALUS 69

nigromaculatus: KU 134092(S). S. praeornatus:
KU 134229%S). S. rhodomelas: KU 152186(S);
USNM 222587(S). S. rosiventris: KU 172196(S);
REE 2284(S).S. varius: KU 121135(S), 134563(S).
Strobilurus: S. torquatus: MCZ 133243(S).
Tapinurus: T. semitaeniatus: REE 1801(S).
Tropidurus: T. atacamensis: KU 161983(S). T.
barringtonensis: SDSNH 3092S). T. etheridgei:
KU 186113(S). T. hispidus: KU 167508(S). T. ko-
epckeorum: KU 163604, 163606, 163607. T. oc-
cipitalis: WP 547, 556. T. peruvianus: KU
134674(S), 164055(S), 164056(S). 7. spinulosus:

KU 97856(S); REE 2470. T. stolzmani: KU 134701,
134706, 134708, 134726. T. theresoides: KU
KGZOU2(S)S 1620 lS (S\eT. thoracieuss KU
163724(S). T. torquatus: REE 324(S); USNM
222582(S).

Uranoscodon: U. superciliaris: KU 135269(S).
U. superciliosa: MAN 44(S); REE 2508, 2511(S).

Urocentron: U. flaviceps: REE 924.

Vilcunia: V. periglacialis: MCZ 122010,
162007, 162008(S), 162009. V. sylvanae: MCZ
156906(S).

it "wy

—
a _
O -eR@asanyt -
q
ES, MATION tie
’ 1a inher aa
a4 :
y
=
' t
,
i
é , }
> aot ht

Harvard MCZ Library AMP 42

70.

Thy

ve.

iy:

74.

ae.

76.

we.

TB.

To:

80.

81.

82.

83.

ATIONS

Seauarsize dy i 7-30-72 2, 9 figures. 27 February 1981.
Paper bound. © tub

Late Pleistocei Pregill. Pp. 1-72, 26 figures.

8 May 1981.

ades of northern Ecuador and
uly 1981. Paper bound.

Leptodactylid |
adjacent Colo

|
m of The University of Kansas
ze, J. D. Stewart, A. M. Neuner

tral and northern Great Plains.
res. 1 June 1983. Paper bound.

Relationships o
By Lawrence

genus Stefania. By William E.
larch 1984. Paper bound.

The taxonomy
Duellman and |

Variation in clu__ eT Ue ii y S. Fitch. Pp. 1-76, 15 figures.
24 May 1985. Paper bound.
Type and figured specimens of fossil vertebrates in the collection of The University of Kansas
Museum of Natural History. Part II. Fossil amphibians and reptiles. By H.-P. Schultze, L. Hunt,
J. Chorn and A. M. Neuner. Pp. 1-66. 3 December 1985. Paper bound.

Type and figured specimens of fossil vertebrates in the collection of The University of Kansas
Museum of Natural History. Part III. Fossil birds. By John F. Neas and Marion Anne Jenkinson.
Pp. 1-14. 5 February 1986. Paper bound.

Type and figured specimens of fossil vertebrates in the collection of The University of Kansas
Museum of Natural History. Part IV. Fossil mammals. By Gregg E. Ostrander, Assefa Mebrate
and Robert W. Wilson. Pp. 1-83. 21 November 1986. Paper bound.

Phylogenetic studies of north american minnows, with emphasis on the genus Cyprinella
(Teleostei: Cypriniformes). By Richard L. Mayden. Pp. 1-189, 85 figures, 4 color plates. 1 June
1989. Paper bound. ISBN: 0-89338—029-6.

A Phylogenetic Analysis and Taxonomy of Iguanian Lizards (Reptilia: Squamata). By Darrel R.
Frost and Richard Etheridge. Pp. 1-65, 24 figures, 3 appendices. 28 September 1989. Paper
bound. ISBN: 0-89338—033-4.

Bats of Portugal: Zoogeography and Systematics. By Jorge M. Palmeirim. Pp. 1—53, 39 figures,
24 tables, 1 appendix. 15 March 1990. Paper bound. ISBN: 0-89338—034-2.

Altitudinal Ecology of Agama tuberculata Gray in the Western Himalayas. By Robert C. Waltner.
Pp. 1-74, 38 figures, 24 tables. 20 February 1991. Paper bound. ISBN: 0—89338—036-9.